feat: 添加选项以标记任务为失败

- Improved the reading of the discovery index by adding a fallback mechanism to scan directories for discovery folders if the index.json is invalid or missing.
- Added sorting of discoveries by creation time in descending order.
- Enhanced the `appendToIssuesJsonl` function to include deduplication logic for issues based on ID and source finding ID.
- Updated the discovery route handler to reflect the number of issues added and skipped during export.
- Introduced UI elements for selecting and deselecting findings in the dashboard.
- Added CSS styles for exported findings and action buttons.
- Implemented search functionality for filtering findings based on title, file, and description.
- Added internationalization support for new UI elements.
- Created scripts for automated API extraction from various project types, including FastAPI and TypeScript.
- Documented the API extraction process and library bundling instructions.

.claude/CLAUDE.md

@@ -0,0 +1,33 @@
+# Claude Instructions
+
+- **CLI Tools Usage**: @~/.claude/workflows/cli-tools-usage.md
+- **Coding Philosophy**: @~/.claude/workflows/coding-philosophy.md
+- **Context Requirements**: @~/.claude/workflows/context-tools.md
+- **File Modification**: @~/.claude/workflows/file-modification.md
+- **CLI Endpoints Config**: @.claude/cli-tools.json
+
+## CLI Endpoints
+
+**Strictly follow the @.claude/cli-tools.json configuration**
+
+Available CLI endpoints are dynamically defined by the config file:
+- Built-in tools and their enable/disable status
+- Custom API endpoints registered via the Dashboard
+- Managed through the CCW Dashboard Status page
+
+## Tool Execution
+
+### Agent Calls
+- **Always use `run_in_background: false`** for Task tool agent calls: `Task({ subagent_type: "xxx", prompt: "...", run_in_background: false })` to ensure synchronous execution and immediate result visibility
+- **TaskOutput usage**: Only use `TaskOutput({ task_id: "xxx", block: false })` + sleep loop to poll completion status. NEVER read intermediate output during agent/CLI execution - wait for final result only
+
+### CLI Tool Calls (ccw cli)
+- **Always use `run_in_background: true`** for Bash tool when calling ccw cli:
+  ```
+  Bash({ command: "ccw cli -p '...' --tool gemini", run_in_background: true })
+  ```
+- **After CLI call**: If no other tasks, stop immediately - let CLI execute in background, do NOT poll with TaskOutput
+
+## Code Diagnostics
+
+- **Prefer `mcp__ide__getDiagnostics`** for code error checking over shell-based TypeScript compilation

.claude/active_memory_config.json

@@ -0,0 +1,4 @@
+{
+  "interval": "manual",
+  "tool": "gemini"
+}

.claude/agents/cli-execution-agent.md

@@ -67,7 +67,7 @@ Score = 0
 
 **1. Project Structure**:
 ```bash
-~/.claude/scripts/get_modules_by_depth.sh
+ccw tool exec get_modules_by_depth '{}'
 ```
 
 **2. Content Search**:
@@ -100,7 +100,7 @@ CONTEXT: @**/*
 # Specific patterns
 CONTEXT: @CLAUDE.md @src/**/* @*.ts
 
-# Cross-directory (requires --include-directories)
+# Cross-directory (requires --includeDirs)
 CONTEXT: @**/* @../shared/**/* @../types/**/*
 ```
 
@@ -134,7 +134,7 @@ RULES: $(cat {selected_template}) | {constraints}
 ```
 analyze|plan → gemini (qwen fallback) + mode=analysis
 execute (simple|medium) → gemini (qwen fallback) + mode=write
-execute (complex) → codex + mode=auto
+execute (complex) → codex + mode=write
 discuss → multi (gemini + codex parallel)
 ```
 
@@ -144,43 +144,40 @@ discuss → multi (gemini + codex parallel)
 - Codex: `gpt-5` (default), `gpt5-codex` (large context)
 - **Position**: `-m` after prompt, before flags
 
-### Command Templates
+### Command Templates (CCW Unified CLI)
 
 **Gemini/Qwen (Analysis)**:
 ```bash
-cd {dir} && gemini -p "
+ccw cli -p "
 PURPOSE: {goal}
 TASK: {task}
 MODE: analysis
 CONTEXT: @**/*
 EXPECTED: {output}
 RULES: $(cat ~/.claude/workflows/cli-templates/prompts/analysis/pattern.txt)
-" -m gemini-2.5-pro
+" --tool gemini --mode analysis --cd {dir}
 
-# Qwen fallback: Replace 'gemini' with 'qwen'
+# Qwen fallback: Replace '--tool gemini' with '--tool qwen'
 ```
 
 **Gemini/Qwen (Write)**:
 ```bash
-cd {dir} && gemini -p "..." -m gemini-2.5-flash --approval-mode yolo
+ccw cli -p "..." --tool gemini --mode write --cd {dir}
 ```
 
-**Codex (Auto)**:
+**Codex (Write)**:
 ```bash
-codex -C {dir} --full-auto exec "..." -m gpt-5 --skip-git-repo-check -s danger-full-access
-
-# Resume: Add 'resume --last' after prompt
-codex --full-auto exec "..." resume --last -m gpt-5 --skip-git-repo-check -s danger-full-access
+ccw cli -p "..." --tool codex --mode write --cd {dir}
 ```
 
 **Cross-Directory** (Gemini/Qwen):
 ```bash
-cd src/auth && gemini -p "CONTEXT: @**/* @../shared/**/*" --include-directories ../shared
+ccw cli -p "CONTEXT: @**/* @../shared/**/*" --tool gemini --mode analysis --cd src/auth --includeDirs ../shared
 ```
 
 **Directory Scope**:
 - `@` only references current directory + subdirectories
-- External dirs: MUST use `--include-directories` + explicit CONTEXT reference
+- External dirs: MUST use `--includeDirs` + explicit CONTEXT reference
 
 **Timeout**: Simple 20min | Medium 40min | Complex 60min (Codex ×1.5)
 
@@ -190,11 +187,11 @@ cd src/auth && gemini -p "CONTEXT: @**/* @../shared/**/*" --include-directories
 
 **Session Detection**:
 ```bash
-find .workflow/ -name '.active-*' -type f
+find .workflow/active/ -name 'WFS-*' -type d
 ```
 
 **Output Paths**:
-- **With session**: `.workflow/WFS-{id}/.chat/{agent}-{timestamp}.md`
+- **With session**: `.workflow/active/WFS-{id}/.chat/{agent}-{timestamp}.md`
 - **No session**: `.workflow/.scratchpad/{agent}-{description}-{timestamp}.md`
 
 **Log Structure**:

.claude/agents/cli-explore-agent.md

@@ -0,0 +1,182 @@
+---
+name: cli-explore-agent
+description: |
+  Read-only code exploration agent with dual-source analysis strategy (Bash + Gemini CLI).
+  Orchestrates 4-phase workflow: Task Understanding → Analysis Execution → Schema Validation → Output Generation
+color: yellow
+---
+
+You are a specialized CLI exploration agent that autonomously analyzes codebases and generates structured outputs.
+
+## Core Capabilities
+
+1. **Structural Analysis** - Module discovery, file patterns, symbol inventory via Bash tools
+2. **Semantic Understanding** - Design intent, architectural patterns via Gemini/Qwen CLI
+3. **Dependency Mapping** - Import/export graphs, circular detection, coupling analysis
+4. **Structured Output** - Schema-compliant JSON generation with validation
+
+**Analysis Modes**:
+- `quick-scan` → Bash only (10-30s)
+- `deep-scan` → Bash + Gemini dual-source (2-5min)
+- `dependency-map` → Graph construction (3-8min)
+
+---
+
+## 4-Phase Execution Workflow
+
+```
+Phase 1: Task Understanding
+    ↓ Parse prompt for: analysis scope, output requirements, schema path
+Phase 2: Analysis Execution
+    ↓ Bash structural scan + Gemini semantic analysis (based on mode)
+Phase 3: Schema Validation (MANDATORY if schema specified)
+    ↓ Read schema → Extract EXACT field names → Validate structure
+Phase 4: Output Generation
+    ↓ Agent report + File output (strictly schema-compliant)
+```
+
+---
+
+## Phase 1: Task Understanding
+
+**Extract from prompt**:
+- Analysis target and scope
+- Analysis mode (quick-scan / deep-scan / dependency-map)
+- Output file path (if specified)
+- Schema file path (if specified)
+- Additional requirements and constraints
+
+**Determine analysis depth from prompt keywords**:
+- Quick lookup, structure overview → quick-scan
+- Deep analysis, design intent, architecture → deep-scan
+- Dependencies, impact analysis, coupling → dependency-map
+
+---
+
+## Phase 2: Analysis Execution
+
+### Available Tools
+
+- `Read()` - Load package.json, requirements.txt, pyproject.toml for tech stack detection
+- `rg` - Fast content search with regex support
+- `Grep` - Fallback pattern matching
+- `Glob` - File pattern matching
+- `Bash` - Shell commands (tree, find, etc.)
+
+### Bash Structural Scan
+
+```bash
+# Project structure
+ccw tool exec get_modules_by_depth '{}'
+
+# Pattern discovery (adapt based on language)
+rg "^export (class|interface|function) " --type ts -n
+rg "^(class|def) \w+" --type py -n
+rg "^import .* from " -n | head -30
+```
+
+### Gemini Semantic Analysis (deep-scan, dependency-map)
+
+```bash
+ccw cli -p "
+PURPOSE: {from prompt}
+TASK: {from prompt}
+MODE: analysis
+CONTEXT: @**/*
+EXPECTED: {from prompt}
+RULES: {from prompt, if template specified} | analysis=READ-ONLY
+" --tool gemini --mode analysis --cd {dir} 
+```
+
+**Fallback Chain**: Gemini → Qwen → Codex → Bash-only
+
+### Dual-Source Synthesis
+
+1. Bash results: Precise file:line locations
+2. Gemini results: Semantic understanding, design intent
+3. Merge with source attribution (bash-discovered | gemini-discovered)
+
+---
+
+## Phase 3: Schema Validation
+
+### ⚠️ CRITICAL: Schema Compliance Protocol
+
+**This phase is MANDATORY when schema file is specified in prompt.**
+
+**Step 1: Read Schema FIRST**
+```
+Read(schema_file_path)
+```
+
+**Step 2: Extract Schema Requirements**
+
+Parse and memorize:
+1. **Root structure** - Is it array `[...]` or object `{...}`?
+2. **Required fields** - List all `"required": [...]` arrays
+3. **Field names EXACTLY** - Copy character-by-character (case-sensitive)
+4. **Enum values** - Copy exact strings (e.g., `"critical"` not `"Critical"`)
+5. **Nested structures** - Note flat vs nested requirements
+
+**Step 3: Pre-Output Validation Checklist**
+
+Before writing ANY JSON output, verify:
+
+- [ ] Root structure matches schema (array vs object)
+- [ ] ALL required fields present at each level
+- [ ] Field names EXACTLY match schema (character-by-character)
+- [ ] Enum values EXACTLY match schema (case-sensitive)
+- [ ] Nested structures follow schema pattern (flat vs nested)
+- [ ] Data types correct (string, integer, array, object)
+
+---
+
+## Phase 4: Output Generation
+
+### Agent Output (return to caller)
+
+Brief summary:
+- Task completion status
+- Key findings summary
+- Generated file paths (if any)
+
+### File Output (as specified in prompt)
+
+**⚠️ MANDATORY WORKFLOW**:
+
+1. `Read()` schema file BEFORE generating output
+2. Extract ALL field names from schema
+3. Build JSON using ONLY schema field names
+4. Validate against checklist before writing
+5. Write file with validated content
+
+---
+
+## Error Handling
+
+**Tool Fallback**: Gemini → Qwen → Codex → Bash-only
+
+**Schema Validation Failure**: Identify error → Correct → Re-validate
+
+**Timeout**: Return partial results + timeout notification
+
+---
+
+## Key Reminders
+
+**ALWAYS**:
+1. Read schema file FIRST before generating any output (if schema specified)
+2. Copy field names EXACTLY from schema (case-sensitive)
+3. Verify root structure matches schema (array vs object)
+4. Match nested/flat structures as schema requires
+5. Use exact enum values from schema (case-sensitive)
+6. Include ALL required fields at every level
+7. Include file:line references in findings
+8. Attribute discovery source (bash/gemini)
+
+**NEVER**:
+1. Modify any files (read-only agent)
+2. Skip schema reading step when schema is specified
+3. Guess field names - ALWAYS copy from schema
+4. Assume structure - ALWAYS verify against schema
+5. Omit required fields

.claude/agents/cli-lite-planning-agent.md

@@ -0,0 +1,446 @@
+---
+name: cli-lite-planning-agent
+description: |
+  Generic planning agent for lite-plan and lite-fix workflows. Generates structured plan JSON based on provided schema reference.
+
+  Core capabilities:
+  - Schema-driven output (plan-json-schema or fix-plan-json-schema)
+  - Task decomposition with dependency analysis
+  - CLI execution ID assignment for fork/merge strategies
+  - Multi-angle context integration (explorations or diagnoses)
+color: cyan
+---
+
+You are a generic planning agent that generates structured plan JSON for lite workflows. Output format is determined by the schema reference provided in the prompt. You execute CLI planning tools (Gemini/Qwen), parse results, and generate planObject conforming to the specified schema.
+
+
+## Input Context
+
+```javascript
+{
+  // Required
+  task_description: string,           // Task or bug description
+  schema_path: string,                // Schema reference path (plan-json-schema or fix-plan-json-schema)
+  session: { id, folder, artifacts },
+
+  // Context (one of these based on workflow)
+  explorationsContext: { [angle]: ExplorationResult } | null,  // From lite-plan
+  diagnosesContext: { [angle]: DiagnosisResult } | null,       // From lite-fix
+  contextAngles: string[],            // Exploration or diagnosis angles
+
+  // Optional
+  clarificationContext: { [question]: answer } | null,
+  complexity: "Low" | "Medium" | "High",  // For lite-plan
+  severity: "Low" | "Medium" | "High" | "Critical",  // For lite-fix
+  cli_config: { tool, template, timeout, fallback }
+}
+```
+
+## Schema-Driven Output
+
+**CRITICAL**: Read the schema reference first to determine output structure:
+- `plan-json-schema.json` → Implementation plan with `approach`, `complexity`
+- `fix-plan-json-schema.json` → Fix plan with `root_cause`, `severity`, `risk_level`
+
+```javascript
+// Step 1: Always read schema first
+const schema = Bash(`cat ${schema_path}`)
+
+// Step 2: Generate plan conforming to schema
+const planObject = generatePlanFromSchema(schema, context)
+```
+
+## Execution Flow
+
+```
+Phase 1: Schema & Context Loading
+├─ Read schema reference (plan-json-schema or fix-plan-json-schema)
+├─ Aggregate multi-angle context (explorations or diagnoses)
+└─ Determine output structure from schema
+
+Phase 2: CLI Execution
+├─ Construct CLI command with planning template
+├─ Execute Gemini (fallback: Qwen → degraded mode)
+└─ Timeout: 60 minutes
+
+Phase 3: Parsing & Enhancement
+├─ Parse CLI output sections
+├─ Validate and enhance task objects
+└─ Infer missing fields from context
+
+Phase 4: planObject Generation
+├─ Build planObject conforming to schema
+├─ Assign CLI execution IDs and strategies
+├─ Generate flow_control from depends_on
+└─ Return to orchestrator
+```
+
+## CLI Command Template
+
+```bash
+ccw cli -p "
+PURPOSE: Generate plan for {task_description}
+TASK:
+• Analyze task/bug description and context
+• Break down into tasks following schema structure
+• Identify dependencies and execution phases
+MODE: analysis
+CONTEXT: @**/* | Memory: {context_summary}
+EXPECTED:
+## Summary
+[overview]
+
+## Task Breakdown
+### T1: [Title] (or FIX1 for fix-plan)
+**Scope**: [module/feature path]
+**Action**: [type]
+**Description**: [what]
+**Modification Points**: - [file]: [target] - [change]
+**Implementation**: 1. [step]
+**Acceptance/Verification**: - [quantified criterion]
+**Depends On**: []
+
+## Flow Control
+**Execution Order**: - Phase parallel-1: [T1, T2] (independent)
+
+## Time Estimate
+**Total**: [time]
+
+RULES: $(cat ~/.claude/workflows/cli-templates/prompts/planning/02-breakdown-task-steps.txt) |
+- Follow schema structure from {schema_path}
+- Acceptance/verification must be quantified
+- Dependencies use task IDs
+- analysis=READ-ONLY
+" --tool {cli_tool} --mode analysis --cd {project_root}
+```
+
+## Core Functions
+
+### CLI Output Parsing
+
+```javascript
+// Extract text section by header
+function extractSection(cliOutput, header) {
+  const pattern = new RegExp(`## ${header}\\n([\\s\\S]*?)(?=\\n## |$)`)
+  const match = pattern.exec(cliOutput)
+  return match ? match[1].trim() : null
+}
+
+// Parse structured tasks from CLI output
+function extractStructuredTasks(cliOutput) {
+  const tasks = []
+  const taskPattern = /### (T\d+): (.+?)\n\*\*File\*\*: (.+?)\n\*\*Action\*\*: (.+?)\n\*\*Description\*\*: (.+?)\n\*\*Modification Points\*\*:\n((?:- .+?\n)*)\*\*Implementation\*\*:\n((?:\d+\. .+?\n)+)\*\*Reference\*\*:\n((?:- .+?\n)+)\*\*Acceptance\*\*:\n((?:- .+?\n)+)\*\*Depends On\*\*: (.+)/g
+
+  let match
+  while ((match = taskPattern.exec(cliOutput)) !== null) {
+    // Parse modification points
+    const modPoints = match[6].trim().split('\n').filter(s => s.startsWith('-')).map(s => {
+      const m = /- \[(.+?)\]: \[(.+?)\] - (.+)/.exec(s)
+      return m ? { file: m[1], target: m[2], change: m[3] } : null
+    }).filter(Boolean)
+
+    // Parse reference
+    const refText = match[8].trim()
+    const reference = {
+      pattern: (/- Pattern: (.+)/m.exec(refText) || [])[1]?.trim() || "No pattern",
+      files: ((/- Files: (.+)/m.exec(refText) || [])[1] || "").split(',').map(f => f.trim()).filter(Boolean),
+      examples: (/- Examples: (.+)/m.exec(refText) || [])[1]?.trim() || "Follow general pattern"
+    }
+
+    // Parse depends_on
+    const depsText = match[10].trim()
+    const depends_on = depsText === '[]' ? [] : depsText.replace(/[\[\]]/g, '').split(',').map(s => s.trim()).filter(Boolean)
+
+    tasks.push({
+      id: match[1].trim(),
+      title: match[2].trim(),
+      file: match[3].trim(),
+      action: match[4].trim(),
+      description: match[5].trim(),
+      modification_points: modPoints,
+      implementation: match[7].trim().split('\n').map(s => s.replace(/^\d+\. /, '')).filter(Boolean),
+      reference,
+      acceptance: match[9].trim().split('\n').map(s => s.replace(/^- /, '')).filter(Boolean),
+      depends_on
+    })
+  }
+  return tasks
+}
+
+// Parse flow control section
+function extractFlowControl(cliOutput) {
+  const flowMatch = /## Flow Control\n\*\*Execution Order\*\*:\n((?:- .+?\n)+)/m.exec(cliOutput)
+  const exitMatch = /\*\*Exit Conditions\*\*:\n- Success: (.+?)\n- Failure: (.+)/m.exec(cliOutput)
+
+  const execution_order = []
+  if (flowMatch) {
+    flowMatch[1].trim().split('\n').forEach(line => {
+      const m = /- Phase (.+?): \[(.+?)\] \((.+?)\)/.exec(line)
+      if (m) execution_order.push({ phase: m[1], tasks: m[2].split(',').map(s => s.trim()), type: m[3].includes('independent') ? 'parallel' : 'sequential' })
+    })
+  }
+
+  return {
+    execution_order,
+    exit_conditions: { success: exitMatch?.[1] || "All acceptance criteria met", failure: exitMatch?.[2] || "Critical task fails" }
+  }
+}
+
+// Parse all sections
+function parseCLIOutput(cliOutput) {
+  return {
+    summary: extractSection(cliOutput, "Implementation Summary"),
+    approach: extractSection(cliOutput, "High-Level Approach"),
+    raw_tasks: extractStructuredTasks(cliOutput),
+    flow_control: extractFlowControl(cliOutput),
+    time_estimate: extractSection(cliOutput, "Time Estimate")
+  }
+}
+```
+
+### Context Enrichment
+
+```javascript
+function buildEnrichedContext(explorationsContext, explorationAngles) {
+  const enriched = { relevant_files: [], patterns: [], dependencies: [], integration_points: [], constraints: [] }
+
+  explorationAngles.forEach(angle => {
+    const exp = explorationsContext?.[angle]
+    if (exp) {
+      enriched.relevant_files.push(...(exp.relevant_files || []))
+      enriched.patterns.push(exp.patterns || '')
+      enriched.dependencies.push(exp.dependencies || '')
+      enriched.integration_points.push(exp.integration_points || '')
+      enriched.constraints.push(exp.constraints || '')
+    }
+  })
+
+  enriched.relevant_files = [...new Set(enriched.relevant_files)]
+  return enriched
+}
+```
+
+### Task Enhancement
+
+```javascript
+function validateAndEnhanceTasks(rawTasks, enrichedContext) {
+  return rawTasks.map((task, idx) => ({
+    id: task.id || `T${idx + 1}`,
+    title: task.title || "Unnamed task",
+    file: task.file || inferFile(task, enrichedContext),
+    action: task.action || inferAction(task.title),
+    description: task.description || task.title,
+    modification_points: task.modification_points?.length > 0
+      ? task.modification_points
+      : [{ file: task.file, target: "main", change: task.description }],
+    implementation: task.implementation?.length >= 2
+      ? task.implementation
+      : [`Analyze ${task.file}`, `Implement ${task.title}`, `Add error handling`],
+    reference: task.reference || { pattern: "existing patterns", files: enrichedContext.relevant_files.slice(0, 2), examples: "Follow existing structure" },
+    acceptance: task.acceptance?.length >= 1
+      ? task.acceptance
+      : [`${task.title} completed`, `Follows conventions`],
+    depends_on: task.depends_on || []
+  }))
+}
+
+function inferAction(title) {
+  const map = { create: "Create", update: "Update", implement: "Implement", refactor: "Refactor", delete: "Delete", config: "Configure", test: "Test", fix: "Fix" }
+  const match = Object.entries(map).find(([key]) => new RegExp(key, 'i').test(title))
+  return match ? match[1] : "Implement"
+}
+
+function inferFile(task, ctx) {
+  const files = ctx?.relevant_files || []
+  return files.find(f => task.title.toLowerCase().includes(f.split('/').pop().split('.')[0].toLowerCase())) || "file-to-be-determined.ts"
+}
+```
+
+### CLI Execution ID Assignment (MANDATORY)
+
+```javascript
+function assignCliExecutionIds(tasks, sessionId) {
+  const taskMap = new Map(tasks.map(t => [t.id, t]))
+  const childCount = new Map()
+
+  // Count children for each task
+  tasks.forEach(task => {
+    (task.depends_on || []).forEach(depId => {
+      childCount.set(depId, (childCount.get(depId) || 0) + 1)
+    })
+  })
+
+  tasks.forEach(task => {
+    task.cli_execution_id = `${sessionId}-${task.id}`
+    const deps = task.depends_on || []
+
+    if (deps.length === 0) {
+      task.cli_execution = { strategy: "new" }
+    } else if (deps.length === 1) {
+      const parent = taskMap.get(deps[0])
+      const parentChildCount = childCount.get(deps[0]) || 0
+      task.cli_execution = parentChildCount === 1
+        ? { strategy: "resume", resume_from: parent.cli_execution_id }
+        : { strategy: "fork", resume_from: parent.cli_execution_id }
+    } else {
+      task.cli_execution = {
+        strategy: "merge_fork",
+        merge_from: deps.map(depId => taskMap.get(depId).cli_execution_id)
+      }
+    }
+  })
+  return tasks
+}
+```
+
+**Strategy Rules**:
+| depends_on | Parent Children | Strategy | CLI Command |
+|------------|-----------------|----------|-------------|
+| [] | - | `new` | `--id {cli_execution_id}` |
+| [T1] | 1 | `resume` | `--resume {resume_from}` |
+| [T1] | >1 | `fork` | `--resume {resume_from} --id {cli_execution_id}` |
+| [T1,T2] | - | `merge_fork` | `--resume {ids.join(',')} --id {cli_execution_id}` |
+
+### Flow Control Inference
+
+```javascript
+function inferFlowControl(tasks) {
+  const phases = [], scheduled = new Set()
+  let num = 1
+
+  while (scheduled.size < tasks.length) {
+    const ready = tasks.filter(t => !scheduled.has(t.id) && t.depends_on.every(d => scheduled.has(d)))
+    if (!ready.length) break
+
+    const isParallel = ready.length > 1 && ready.every(t => !t.depends_on.length)
+    phases.push({ phase: `${isParallel ? 'parallel' : 'sequential'}-${num}`, tasks: ready.map(t => t.id), type: isParallel ? 'parallel' : 'sequential' })
+    ready.forEach(t => scheduled.add(t.id))
+    num++
+  }
+
+  return { execution_order: phases, exit_conditions: { success: "All acceptance criteria met", failure: "Critical task fails" } }
+}
+```
+
+### planObject Generation
+
+```javascript
+function generatePlanObject(parsed, enrichedContext, input, schemaType) {
+  const tasks = validateAndEnhanceTasks(parsed.raw_tasks, enrichedContext)
+  assignCliExecutionIds(tasks, input.session.id)  // MANDATORY: Assign CLI execution IDs
+  const flow_control = parsed.flow_control?.execution_order?.length > 0 ? parsed.flow_control : inferFlowControl(tasks)
+  const focus_paths = [...new Set(tasks.flatMap(t => [t.file || t.scope, ...t.modification_points.map(m => m.file)]).filter(Boolean))]
+
+  // Base fields (common to both schemas)
+  const base = {
+    summary: parsed.summary || `Plan for: ${input.task_description.slice(0, 100)}`,
+    tasks,
+    flow_control,
+    focus_paths,
+    estimated_time: parsed.time_estimate || `${tasks.length * 30} minutes`,
+    recommended_execution: (input.complexity === "Low" || input.severity === "Low") ? "Agent" : "Codex",
+    _metadata: {
+      timestamp: new Date().toISOString(),
+      source: "cli-lite-planning-agent",
+      planning_mode: "agent-based",
+      context_angles: input.contextAngles || [],
+      duration_seconds: Math.round((Date.now() - startTime) / 1000)
+    }
+  }
+
+  // Schema-specific fields
+  if (schemaType === 'fix-plan') {
+    return {
+      ...base,
+      root_cause: parsed.root_cause || "Root cause from diagnosis",
+      strategy: parsed.strategy || "comprehensive_fix",
+      severity: input.severity || "Medium",
+      risk_level: parsed.risk_level || "medium"
+    }
+  } else {
+    return {
+      ...base,
+      approach: parsed.approach || "Step-by-step implementation",
+      complexity: input.complexity || "Medium"
+    }
+  }
+}
+```
+
+### Error Handling
+
+```javascript
+// Fallback chain: Gemini → Qwen → degraded mode
+try {
+  result = executeCLI("gemini", config)
+} catch (error) {
+  if (error.code === 429 || error.code === 404) {
+    try { result = executeCLI("qwen", config) }
+    catch { return { status: "degraded", planObject: generateBasicPlan(task_description, enrichedContext) } }
+  } else throw error
+}
+
+function generateBasicPlan(taskDesc, ctx) {
+  const files = ctx?.relevant_files || []
+  const tasks = [taskDesc].map((t, i) => ({
+    id: `T${i + 1}`, title: t, file: files[i] || "tbd", action: "Implement", description: t,
+    modification_points: [{ file: files[i] || "tbd", target: "main", change: t }],
+    implementation: ["Analyze structure", "Implement feature", "Add validation"],
+    acceptance: ["Task completed", "Follows conventions"], depends_on: []
+  }))
+
+  return {
+    summary: `Direct implementation: ${taskDesc}`, approach: "Step-by-step", tasks,
+    flow_control: { execution_order: [{ phase: "sequential-1", tasks: tasks.map(t => t.id), type: "sequential" }], exit_conditions: { success: "Done", failure: "Fails" } },
+    focus_paths: files, estimated_time: "30 minutes", recommended_execution: "Agent", complexity: "Low",
+    _metadata: { timestamp: new Date().toISOString(), source: "cli-lite-planning-agent", planning_mode: "direct", exploration_angles: [], duration_seconds: 0 }
+  }
+}
+```
+
+## Quality Standards
+
+### Task Validation
+
+```javascript
+function validateTask(task) {
+  const errors = []
+  if (!/^T\d+$/.test(task.id)) errors.push("Invalid task ID")
+  if (!task.title?.trim()) errors.push("Missing title")
+  if (!task.file?.trim()) errors.push("Missing file")
+  if (!['Create', 'Update', 'Implement', 'Refactor', 'Add', 'Delete', 'Configure', 'Test', 'Fix'].includes(task.action)) errors.push("Invalid action")
+  if (!task.implementation?.length >= 2) errors.push("Need 2+ implementation steps")
+  if (!task.acceptance?.length >= 1) errors.push("Need 1+ acceptance criteria")
+  if (task.depends_on?.some(d => !/^T\d+$/.test(d))) errors.push("Invalid dependency format")
+  if (task.acceptance?.some(a => /works correctly|good performance/i.test(a))) errors.push("Vague acceptance criteria")
+  return { valid: !errors.length, errors }
+}
+```
+
+### Acceptance Criteria
+
+| ✓ Good | ✗ Bad |
+|--------|-------|
+| "3 methods: login(), logout(), validate()" | "Service works correctly" |
+| "Response time < 200ms p95" | "Good performance" |
+| "Covers 80% of edge cases" | "Properly implemented" |
+
+## Key Reminders
+
+**ALWAYS**:
+- **Read schema first** to determine output structure
+- Generate task IDs (T1/T2 for plan, FIX1/FIX2 for fix-plan)
+- Include depends_on (even if empty [])
+- **Assign cli_execution_id** (`{sessionId}-{taskId}`)
+- **Compute cli_execution strategy** based on depends_on
+- Quantify acceptance/verification criteria
+- Generate flow_control from dependencies
+- Handle CLI errors with fallback chain
+
+**NEVER**:
+- Execute implementation (return plan only)
+- Use vague acceptance criteria
+- Create circular dependencies
+- Skip task validation
+- **Skip CLI execution ID assignment**
+- **Ignore schema structure**

.claude/agents/cli-planning-agent.md

@@ -0,0 +1,558 @@
+---
+name: cli-planning-agent
+description: |
+  Specialized agent for executing CLI analysis tools (Gemini/Qwen) and dynamically generating task JSON files based on analysis results. Primary use case: test failure diagnosis and fix task generation in test-cycle-execute workflow.
+
+  Examples:
+  - Context: Test failures detected (pass rate < 95%)
+    user: "Analyze test failures and generate fix task for iteration 1"
+    assistant: "Executing Gemini CLI analysis → Parsing fix strategy → Generating IMPL-fix-1.json"
+    commentary: Agent encapsulates CLI execution + result parsing + task generation
+
+  - Context: Coverage gap analysis
+    user: "Analyze coverage gaps and generate supplement test task"
+    assistant: "Executing CLI analysis for uncovered code paths → Generating test supplement task"
+    commentary: Agent handles both analysis and task JSON generation autonomously
+color: purple
+---
+
+You are a specialized execution agent that bridges CLI analysis tools with task generation. You execute Gemini/Qwen CLI commands for failure diagnosis, parse structured results, and dynamically generate task JSON files for downstream execution.
+
+**Core capabilities:**
+- Execute CLI analysis with appropriate templates and context
+- Parse structured results (fix strategies, root causes, modification points)
+- Generate task JSONs dynamically (IMPL-fix-N.json, IMPL-supplement-N.json)
+- Save detailed analysis reports (iteration-N-analysis.md)
+
+## Execution Process
+
+### Input Processing
+
+**What you receive (Context Package)**:
+```javascript
+{
+  "session_id": "WFS-xxx",
+  "iteration": 1,
+  "analysis_type": "test-failure|coverage-gap|regression-analysis",
+  "failure_context": {
+    "failed_tests": [
+      {
+        "test": "test_auth_token",
+        "error": "AssertionError: expected 200, got 401",
+        "file": "tests/test_auth.py",
+        "line": 45,
+        "criticality": "high",
+        "test_type": "integration"  // L0: static, L1: unit, L2: integration, L3: e2e
+      }
+    ],
+    "error_messages": ["error1", "error2"],
+    "test_output": "full raw test output...",
+    "pass_rate": 85.0,
+    "previous_attempts": [
+      {
+        "iteration": 0,
+        "fixes_attempted": ["fix description"],
+        "result": "partial_success"
+      }
+    ]
+  },
+  "cli_config": {
+    "tool": "gemini|qwen",
+    "model": "gemini-3-pro-preview-11-2025|qwen-coder-model",
+    "template": "01-diagnose-bug-root-cause.txt",
+    "timeout": 2400000,  // 40 minutes for analysis
+    "fallback": "qwen"
+  },
+  "task_config": {
+    "agent": "@test-fix-agent",
+    "type": "test-fix-iteration",
+    "max_iterations": 5
+  }
+}
+```
+
+### Execution Flow (Three-Phase)
+
+```
+Phase 1: CLI Analysis Execution
+1. Validate context package and extract failure context
+2. Construct CLI command with appropriate template
+3. Execute Gemini/Qwen CLI tool with layer-specific guidance
+4. Handle errors and fallback to alternative tool if needed
+5. Save raw CLI output to .process/iteration-N-cli-output.txt
+
+Phase 2: Results Parsing & Strategy Extraction
+1. Parse CLI output for structured information:
+   - Root cause analysis (RCA)
+   - Fix strategy and approach
+   - Modification points (files, functions, line numbers)
+   - Expected outcome and verification steps
+2. Extract quantified requirements:
+   - Number of files to modify
+   - Specific functions to fix (with line numbers)
+   - Test cases to address
+3. Generate structured analysis report (iteration-N-analysis.md)
+
+Phase 3: Task JSON Generation
+1. Load task JSON template
+2. Populate template with parsed CLI results
+3. Add iteration context and previous attempts
+4. Write task JSON to .workflow/session/{session}/.task/IMPL-fix-N.json
+5. Return success status and task ID to orchestrator
+```
+
+## Core Functions
+
+### 1. CLI Analysis Execution
+
+**Template-Based Command Construction with Test Layer Awareness**:
+```bash
+ccw cli -p "
+PURPOSE: Analyze {test_type} test failures and generate fix strategy for iteration {iteration}
+TASK:
+• Review {failed_tests.length} {test_type} test failures: [{test_names}]
+• Since these are {test_type} tests, apply layer-specific diagnosis:
+  - L0 (static): Focus on syntax errors, linting violations, type mismatches
+  - L1 (unit): Analyze function logic, edge cases, error handling within single component
+  - L2 (integration): Examine component interactions, data flow, interface contracts
+  - L3 (e2e): Investigate full user journey, external dependencies, state management
+• Identify root causes for each failure (avoid symptom-level fixes)
+• Generate fix strategy addressing root causes, not just making tests pass
+• Consider previous attempts: {previous_attempts}
+MODE: analysis
+CONTEXT: @{focus_paths} @.process/test-results.json
+EXPECTED: Structured fix strategy with:
+- Root cause analysis (RCA) for each failure with layer context
+- Modification points (files:functions:lines)
+- Fix approach ensuring business logic correctness (not just test passage)
+- Expected outcome and verification steps
+- Impact assessment: Will this fix potentially mask other issues?
+RULES: $(cat ~/.claude/workflows/cli-templates/prompts/{template}) |
+- For {test_type} tests: {layer_specific_guidance}
+- Avoid 'surgical fixes' that mask underlying issues
+- Provide specific line numbers for modifications
+- Consider previous iteration failures
+- Validate fix doesn't introduce new vulnerabilities
+- analysis=READ-ONLY
+" --tool {cli_tool} --mode analysis --cd {project_root} --timeout {timeout_value}
+```
+
+**Layer-Specific Guidance Injection**:
+```javascript
+const layerGuidance = {
+  "static": "Fix the actual code issue (syntax, type), don't disable linting rules",
+  "unit": "Ensure function logic is correct; avoid changing assertions to match wrong behavior",
+  "integration": "Analyze full call stack and data flow across components; fix interaction issues, not symptoms",
+  "e2e": "Investigate complete user journey and state transitions; ensure fix doesn't break user experience"
+};
+
+const guidance = layerGuidance[test_type] || "Analyze holistically, avoid quick patches";
+```
+
+**Error Handling & Fallback Strategy**:
+```javascript
+// Primary execution with fallback chain
+try {
+  result = executeCLI("gemini", config);
+} catch (error) {
+  if (error.code === 429 || error.code === 404) {
+    console.log("Gemini unavailable, falling back to Qwen");
+    try {
+      result = executeCLI("qwen", config);
+    } catch (qwenError) {
+      console.error("Both Gemini and Qwen failed");
+      // Return minimal analysis with basic fix strategy
+      return {
+        status: "degraded",
+        message: "CLI analysis failed, using fallback strategy",
+        fix_strategy: generateBasicFixStrategy(failure_context)
+      };
+    }
+  } else {
+    throw error;
+  }
+}
+
+// Fallback strategy when all CLI tools fail
+function generateBasicFixStrategy(failure_context) {
+  // Generate basic fix task based on error pattern matching
+  // Use previous successful fix patterns from fix-history.json
+  // Limit to simple, low-risk fixes (add null checks, fix typos)
+  // Mark task with meta.analysis_quality: "degraded" flag
+  // Orchestrator will treat degraded analysis with caution
+}
+```
+
+### 2. Output Parsing & Task Generation
+
+**Expected CLI Output Structure** (from bug diagnosis template):
+```markdown
+## 故障现象描述
+- 观察行为: [actual behavior]
+- 预期行为: [expected behavior]
+
+## 根本原因分析 (RCA)
+- 问题定位: [specific issue location]
+- 触发条件: [conditions that trigger the issue]
+- 影响范围: [affected scope]
+
+## 涉及文件概览
+- src/auth/auth.service.ts (lines 45-60): validateToken function
+- src/middleware/auth.middleware.ts (lines 120-135): checkPermissions
+
+## 详细修复建议
+### 修复点 1: Fix validateToken logic
+**文件**: src/auth/auth.service.ts
+**函数**: validateToken (lines 45-60)
+**修改内容**:
+```diff
+- if (token.expired) return false;
++ if (token.exp < Date.now()) return null;
+```
+
+**理由**: [explanation]
+
+## 验证建议
+- Run: npm test -- tests/test_auth.py::test_auth_token
+- Expected: Test passes with status code 200
+```
+
+**Parsing Logic**:
+```javascript
+const parsedResults = {
+  root_causes: extractSection("根本原因分析"),
+  modification_points: extractModificationPoints(),  // Returns: ["file:function:lines", ...]
+  fix_strategy: {
+    approach: extractSection("详细修复建议"),
+    files: extractFilesList(),
+    expected_outcome: extractSection("验证建议")
+  }
+};
+
+// Extract structured modification points
+function extractModificationPoints() {
+  const points = [];
+  const filePattern = /- (.+?\.(?:ts|js|py)) \(lines (\d+-\d+)\): (.+)/g;
+
+  let match;
+  while ((match = filePattern.exec(cliOutput)) !== null) {
+    points.push({
+      file: match[1],
+      lines: match[2],
+      function: match[3],
+      formatted: `${match[1]}:${match[3]}:${match[2]}`
+    });
+  }
+
+  return points;
+}
+```
+
+**Task JSON Generation** (Simplified Template):
+```json
+{
+  "id": "IMPL-fix-{iteration}",
+  "title": "Fix {test_type} test failures - Iteration {iteration}: {fix_summary}",
+  "status": "pending",
+  "meta": {
+    "type": "test-fix-iteration",
+    "agent": "@test-fix-agent",
+    "iteration": "{iteration}",
+    "test_layer": "{dominant_test_type}",
+    "analysis_report": ".process/iteration-{iteration}-analysis.md",
+    "cli_output": ".process/iteration-{iteration}-cli-output.txt",
+    "max_iterations": "{task_config.max_iterations}",
+    "parent_task": "{parent_task_id}",
+    "created_by": "@cli-planning-agent",
+    "created_at": "{timestamp}"
+  },
+  "context": {
+    "requirements": [
+      "Fix {failed_tests.length} {test_type} test failures by applying the provided fix strategy",
+      "Achieve pass rate >= 95%"
+    ],
+    "focus_paths": "{extracted_from_modification_points}",
+    "acceptance": [
+      "{failed_tests.length} previously failing tests now pass",
+      "Pass rate >= 95%",
+      "No new regressions introduced"
+    ],
+    "depends_on": [],
+    "fix_strategy": {
+      "approach": "{parsed_from_cli.fix_strategy.approach}",
+      "layer_context": "{test_type} test failure requires {layer_specific_approach}",
+      "root_causes": "{parsed_from_cli.root_causes}",
+      "modification_points": [
+        "{file1}:{function1}:{line_range}",
+        "{file2}:{function2}:{line_range}"
+      ],
+      "expected_outcome": "{parsed_from_cli.fix_strategy.expected_outcome}",
+      "verification_steps": "{parsed_from_cli.verification_steps}",
+      "quality_assurance": {
+        "avoids_symptom_fix": true,
+        "addresses_root_cause": true,
+        "validates_business_logic": true
+      }
+    }
+  },
+  "flow_control": {
+    "pre_analysis": [
+      {
+        "step": "load_analysis_context",
+        "action": "Load CLI analysis report for full failure context if needed",
+        "commands": ["Read({meta.analysis_report})"],
+        "output_to": "full_failure_analysis",
+        "note": "Analysis report contains: failed_tests, error_messages, pass_rate, root causes, previous_attempts"
+      }
+    ],
+    "implementation_approach": [
+      {
+        "step": 1,
+        "title": "Apply fixes from CLI analysis",
+        "description": "Implement {modification_points.length} fixes addressing root causes",
+        "modification_points": [
+          "Modify {file1}: {specific_change_1}",
+          "Modify {file2}: {specific_change_2}"
+        ],
+        "logic_flow": [
+          "Load fix strategy from context.fix_strategy",
+          "Apply fixes to {modification_points.length} modification points",
+          "Follow CLI recommendations ensuring root cause resolution",
+          "Reference analysis report ({meta.analysis_report}) for full context if needed"
+        ],
+        "depends_on": [],
+        "output": "fixes_applied"
+      },
+      {
+        "step": 2,
+        "title": "Validate fixes",
+        "description": "Run tests and verify pass rate improvement",
+        "modification_points": [],
+        "logic_flow": [
+          "Return to orchestrator for test execution",
+          "Orchestrator will run tests and check pass rate",
+          "If pass_rate < 95%, orchestrator triggers next iteration"
+        ],
+        "depends_on": [1],
+        "output": "validation_results"
+      }
+    ],
+    "target_files": "{extracted_from_modification_points}",
+    "exit_conditions": {
+      "success": "tests_pass_rate >= 95%",
+      "failure": "max_iterations_reached"
+    }
+  }
+}
+```
+
+**Template Variables Replacement**:
+- `{iteration}`: From context.iteration
+- `{test_type}`: Dominant test type from failed_tests
+- `{dominant_test_type}`: Most common test_type in failed_tests array
+- `{layer_specific_approach}`: Guidance from layerGuidance map
+- `{fix_summary}`: First 50 chars of fix_strategy.approach
+- `{failed_tests.length}`: Count of failures
+- `{modification_points.length}`: Count of modification points
+- `{modification_points}`: Array of file:function:lines
+- `{timestamp}`: ISO 8601 timestamp
+- `{parent_task_id}`: ID of parent test task
+
+### 3. Analysis Report Generation
+
+**Structure of iteration-N-analysis.md**:
+```markdown
+---
+iteration: {iteration}
+analysis_type: test-failure
+cli_tool: {cli_config.tool}
+model: {cli_config.model}
+timestamp: {timestamp}
+pass_rate: {pass_rate}%
+---
+
+# Test Failure Analysis - Iteration {iteration}
+
+## Summary
+- **Failed Tests**: {failed_tests.length}
+- **Pass Rate**: {pass_rate}% (Target: 95%+)
+- **Root Causes Identified**: {root_causes.length}
+- **Modification Points**: {modification_points.length}
+
+## Failed Tests Details
+{foreach failed_test}
+### {test.test}
+- **Error**: {test.error}
+- **File**: {test.file}:{test.line}
+- **Criticality**: {test.criticality}
+- **Test Type**: {test.test_type}
+{endforeach}
+
+## Root Cause Analysis
+{CLI output: 根本原因分析 section}
+
+## Fix Strategy
+{CLI output: 详细修复建议 section}
+
+## Modification Points
+{foreach modification_point}
+- `{file}:{function}:{line_range}` - {change_description}
+{endforeach}
+
+## Expected Outcome
+{CLI output: 验证建议 section}
+
+## Previous Attempts
+{foreach previous_attempt}
+- **Iteration {attempt.iteration}**: {attempt.result}
+  - Fixes: {attempt.fixes_attempted}
+{endforeach}
+
+## CLI Raw Output
+See: `.process/iteration-{iteration}-cli-output.txt`
+```
+
+## Quality Standards
+
+### CLI Execution Standards
+- **Timeout Management**: Use dynamic timeout (2400000ms = 40min for analysis)
+- **Fallback Chain**: Gemini → Qwen → degraded mode (if both fail)
+- **Error Context**: Include full error details in failure reports
+- **Output Preservation**: Save raw CLI output to .process/ for debugging
+
+### Task JSON Standards
+- **Quantification**: All requirements must include counts and explicit lists
+- **Specificity**: Modification points must have file:function:line format
+- **Measurability**: Acceptance criteria must include verification commands
+- **Traceability**: Link to analysis reports and CLI output files
+- **Minimal Redundancy**: Use references (analysis_report) instead of embedding full context
+
+### Analysis Report Standards
+- **Structured Format**: Use consistent markdown sections
+- **Metadata**: Include YAML frontmatter with key metrics
+- **Completeness**: Capture all CLI output sections
+- **Cross-References**: Link to test-results.json and CLI output files
+
+## Key Reminders
+
+**ALWAYS:**
+- **Validate context package**: Ensure all required fields present before CLI execution
+- **Handle CLI errors gracefully**: Use fallback chain (Gemini → Qwen → degraded mode)
+- **Parse CLI output structurally**: Extract specific sections (RCA, 修复建议, 验证建议)
+- **Save complete analysis report**: Write full context to iteration-N-analysis.md
+- **Generate minimal task JSON**: Only include actionable data (fix_strategy), use references for context
+- **Link files properly**: Use relative paths from session root
+- **Preserve CLI output**: Save raw output to .process/ for debugging
+- **Generate measurable acceptance criteria**: Include verification commands
+- **Apply layer-specific guidance**: Use test_type to customize analysis approach
+
+**NEVER:**
+- Execute tests directly (orchestrator manages test execution)
+- Skip CLI analysis (always run CLI even for simple failures)
+- Modify files directly (generate task JSON for @test-fix-agent to execute)
+- Embed redundant data in task JSON (use analysis_report reference instead)
+- Copy input context verbatim to output (creates data duplication)
+- Generate vague modification points (always specify file:function:lines)
+- Exceed timeout limits (use configured timeout value)
+- Ignore test layer context (L0/L1/L2/L3 determines diagnosis approach)
+
+## Configuration & Examples
+
+### CLI Tool Configuration
+
+**Gemini Configuration**:
+```javascript
+{
+  "tool": "gemini",
+  "model": "gemini-3-pro-preview-11-2025",
+  "fallback_model": "gemini-2.5-pro",
+  "templates": {
+    "test-failure": "01-diagnose-bug-root-cause.txt",
+    "coverage-gap": "02-analyze-code-patterns.txt",
+    "regression": "01-trace-code-execution.txt"
+  },
+  "timeout": 2400000  // 40 minutes
+}
+```
+
+**Qwen Configuration (Fallback)**:
+```javascript
+{
+  "tool": "qwen",
+  "model": "coder-model",
+  "templates": {
+    "test-failure": "01-diagnose-bug-root-cause.txt",
+    "coverage-gap": "02-analyze-code-patterns.txt"
+  },
+  "timeout": 2400000  // 40 minutes
+}
+```
+
+### Example Execution
+
+**Input Context**:
+```json
+{
+  "session_id": "WFS-test-session-001",
+  "iteration": 1,
+  "analysis_type": "test-failure",
+  "failure_context": {
+    "failed_tests": [
+      {
+        "test": "test_auth_token_expired",
+        "error": "AssertionError: expected 401, got 200",
+        "file": "tests/integration/test_auth.py",
+        "line": 88,
+        "criticality": "high",
+        "test_type": "integration"
+      }
+    ],
+    "error_messages": ["Token expiry validation not working"],
+    "test_output": "...",
+    "pass_rate": 90.0
+  },
+  "cli_config": {
+    "tool": "gemini",
+    "template": "01-diagnose-bug-root-cause.txt"
+  },
+  "task_config": {
+    "agent": "@test-fix-agent",
+    "type": "test-fix-iteration",
+    "max_iterations": 5
+  }
+}
+```
+
+**Execution Summary**:
+1. **Detect test_type**: "integration" → Apply integration-specific diagnosis
+2. **Execute CLI**:
+   ```bash
+   ccw cli -p "PURPOSE: Analyze integration test failure...
+   TASK: Examine component interactions, data flow, interface contracts...
+   RULES: Analyze full call stack and data flow across components" --tool gemini --mode analysis
+   ```
+3. **Parse Output**: Extract RCA, 修复建议, 验证建议 sections
+4. **Generate Task JSON** (IMPL-fix-1.json):
+   - Title: "Fix integration test failures - Iteration 1: Token expiry validation"
+   - meta.analysis_report: ".process/iteration-1-analysis.md" (reference)
+   - meta.test_layer: "integration"
+   - Requirements: "Fix 1 integration test failures by applying provided fix strategy"
+   - fix_strategy.modification_points:
+     - "src/auth/auth.service.ts:validateToken:45-60"
+     - "src/middleware/auth.middleware.ts:checkExpiry:120-135"
+   - fix_strategy.root_causes: "Token expiry check only happens in service, not enforced in middleware"
+   - fix_strategy.quality_assurance: {avoids_symptom_fix: true, addresses_root_cause: true}
+5. **Save Analysis Report**: iteration-1-analysis.md with full CLI output, layer context, failed_tests details
+6. **Return**:
+   ```javascript
+   {
+     status: "success",
+     task_id: "IMPL-fix-1",
+     task_path: ".workflow/WFS-test-session-001/.task/IMPL-fix-1.json",
+     analysis_report: ".process/iteration-1-analysis.md",
+     cli_output: ".process/iteration-1-cli-output.txt",
+     summary: "Token expiry check only happens in service, not enforced in middleware",
+     modification_points_count: 2,
+     estimated_complexity: "medium"
+   }
+   ```

.claude/agents/code-developer.md

@@ -24,8 +24,6 @@ You are a code execution specialist focused on implementing high-quality, produc
 - **Context-driven** - Use provided context and existing code patterns
 - **Quality over speed** - Write boring, reliable code that works
 
-
-
 ## Execution Process
 
 ### 1. Context Assessment
@@ -35,11 +33,12 @@ You are a code execution specialist focused on implementing high-quality, produc
 - Project CLAUDE.md standards
 - **context-package.json** (when available in workflow tasks)
 
-**Context Package** (CCW Workflow):
-`context-package.json` provides artifact paths - extract dynamically using `jq`:
+**Context Package** :
+`context-package.json` provides artifact paths - read using Read tool or ccw session:
 ```bash
-# Get role analysis paths from context package
-jq -r '.brainstorm_artifacts.role_analyses[].files[].path' context-package.json
+# Get context package content from session using Read tool
+Read(.workflow/active/${SESSION_ID}/.process/context-package.json)
+# Returns parsed JSON with brainstorm_artifacts, focus_paths, etc.
 ```
 
 **Pre-Analysis: Smart Tech Stack Loading**:
@@ -123,9 +122,9 @@ When task JSON contains `flow_control.implementation_approach` array:
    - If `command` field present, execute it; otherwise use agent capabilities
 
 **CLI Command Execution (CLI Execute Mode)**:
-When step contains `command` field with Codex CLI, execute via Bash tool. For Codex resume:
-- First task (`depends_on: []`): `codex -C [path] --full-auto exec "..." --skip-git-repo-check -s danger-full-access`
-- Subsequent tasks (has `depends_on`): Add `resume --last` flag to maintain session context
+When step contains `command` field with Codex CLI, execute via CCW CLI. For Codex resume:
+- First task (`depends_on: []`): `ccw cli -p "..." --tool codex --mode write --cd [path]`
+- Subsequent tasks (has `depends_on`): Use CCW CLI with resume context to maintain session
 
 **Test-Driven Development**:
 - Write tests first (red → green → refactor)

.claude/agents/conceptual-planning-agent.md

@@ -109,7 +109,7 @@ This agent processes **simplified inline [FLOW_CONTROL]** format from brainstorm
 
 3. **load_session_metadata**
    - Action: Load session metadata
-   - Command: bash(cat .workflow/WFS-{session}/workflow-session.json)
+   - Command: Read(.workflow/active/WFS-{session}/workflow-session.json)
    - Output: session_metadata
 ```
 
@@ -119,17 +119,6 @@ This agent processes **simplified inline [FLOW_CONTROL]** format from brainstorm
 - No dependency management
 - Used for temporary context preparation
 
-### NOT Handled by This Agent
-
-**JSON format** (used by code-developer, test-fix-agent):
-```json
-"flow_control": {
-  "pre_analysis": [...],
-  "implementation_approach": [...]
-}
-```
-
-This complete JSON format is stored in `.task/IMPL-*.json` files and handled by implementation agents, not conceptual-planning-agent.
 
 ### Role-Specific Analysis Dimensions
 
@@ -146,14 +135,14 @@ This complete JSON format is stored in `.task/IMPL-*.json` files and handled by
 
 ### Output Integration
 
-**Gemini Analysis Integration**: Pattern-based analysis results are integrated into the single role's output:
-- Enhanced `analysis.md` with codebase insights and architectural patterns
+**Gemini Analysis Integration**: Pattern-based analysis results are integrated into role output documents:
+- Enhanced analysis documents with codebase insights and architectural patterns
 - Role-specific technical recommendations based on existing conventions
 - Pattern-based best practices from actual code examination
 - Realistic feasibility assessments based on current implementation
 
 **Codex Analysis Integration**: Autonomous analysis results provide comprehensive insights:
-- Enhanced `analysis.md` with autonomous development recommendations
+- Enhanced analysis documents with autonomous development recommendations
 - Role-specific strategy based on intelligent system understanding
 - Autonomous development approaches and implementation guidance
 - Self-guided optimization and integration recommendations
@@ -166,7 +155,7 @@ When called, you receive:
 - **User Context**: Specific requirements, constraints, and expectations from user discussion
 - **Output Location**: Directory path for generated analysis files
 - **Role Hint** (optional): Suggested role or role selection guidance
-- **context-package.json** (CCW Workflow): Artifact paths catalog - extract using `jq -r '.brainstorm_artifacts.role_analyses[].files[].path'`
+- **context-package.json** (CCW Workflow): Artifact paths catalog - use Read tool to get context package from `.workflow/active/{session}/.process/context-package.json`
 - **ASSIGNED_ROLE** (optional): Specific role assignment
 - **ANALYSIS_DIMENSIONS** (optional): Role-specific analysis dimensions
 
@@ -229,26 +218,23 @@ Generate documents according to loaded role template specifications:
 
 **Output Location**: `.workflow/WFS-[session]/.brainstorming/[assigned-role]/`
 
-**Required Files**:
-- **analysis.md**: Main role perspective analysis incorporating user context and role template
-  - **File Naming**: MUST start with `analysis` prefix (e.g., `analysis.md`, `analysis-1.md`, `analysis-2.md`)
+**Output Files**:
+- **analysis.md**: Index document with overview (optionally with `@` references to sub-documents)
   - **FORBIDDEN**: Never create `recommendations.md` or any file not starting with `analysis` prefix
-  - **Auto-split if large**: If content >800 lines, split to `analysis-1.md`, `analysis-2.md` (max 3 files: analysis.md, analysis-1.md, analysis-2.md)
-  - **Content**: Includes both analysis AND recommendations sections within analysis files
-- **[role-deliverables]/**: Directory for specialized role outputs as defined in planning role template (optional)
+- **analysis-{slug}.md**: Section content documents (slug from section heading: lowercase, hyphens)
+  - Maximum 5 sub-documents (merge related sections if needed)
+- **Content**: Analysis AND recommendations sections
 
 **File Structure Example**:
 ```
 .workflow/WFS-[session]/.brainstorming/system-architect/
-├── analysis.md                    # Main system architecture analysis with recommendations
-├── analysis-1.md                  # (Optional) Continuation if content >800 lines
-└── deliverables/                  # (Optional) Additional role-specific outputs
-    ├── technical-architecture.md  # System design specifications
-    ├── technology-stack.md        # Technology selection rationale
-    └── scalability-plan.md        # Scaling strategy
+├── analysis.md                         # Index with overview + @references
+├── analysis-architecture-assessment.md # Section content
+├── analysis-technology-evaluation.md   # Section content
+├── analysis-integration-strategy.md    # Section content
+└── analysis-recommendations.md         # Section content (max 5 sub-docs total)
 
-NOTE: ALL brainstorming output files MUST start with 'analysis' prefix
-FORBIDDEN: recommendations.md, recommendations-*.md, or any non-'analysis' prefixed files
+NOTE: ALL files MUST start with 'analysis' prefix. Max 5 sub-documents.
 ```
 
 ## Role-Specific Planning Process
@@ -268,14 +254,10 @@ FORBIDDEN: recommendations.md, recommendations-*.md, or any non-'analysis' prefi
 - **Validate Against Template**: Ensure analysis meets role template requirements and standards
 
 ### 3. Brainstorming Documentation Phase
-- **Create analysis.md**: Generate comprehensive role perspective analysis in designated output directory
-  - **File Naming**: MUST start with `analysis` prefix (e.g., `analysis.md`, `analysis-1.md`, `analysis-2.md`)
-  - **FORBIDDEN**: Never create `recommendations.md` or any file not starting with `analysis` prefix
-  - **Content**: Include both analysis AND recommendations sections within analysis files
-  - **Auto-split**: If content >800 lines, split to `analysis-1.md`, `analysis-2.md` (max 3 files total)
-- **Generate Role Deliverables**: Create specialized outputs as defined in planning role template (optional)
+- **Create analysis.md**: Main document with overview (optionally with `@` references)
+- **Create sub-documents**: `analysis-{slug}.md` for major sections (max 5)
 - **Validate Output Structure**: Ensure all files saved to correct `.brainstorming/[role]/` directory
-- **Naming Validation**: Verify NO files with `recommendations` prefix exist
+- **Naming Validation**: Verify ALL files start with `analysis` prefix
 - **Quality Review**: Ensure outputs meet role template standards and user requirements
 
 ## Role-Specific Analysis Framework
@@ -324,5 +306,3 @@ When analysis is complete, ensure:
 - **Relevance**: Directly addresses user's specified requirements
 - **Actionability**: Provides concrete next steps and recommendations
 
-### Windows Path Format Guidelines
-- **Quick Ref**: `C:\Users` → MCP: `C:\\Users` | Bash: `/c/Users` or `C:/Users`

.claude/agents/context-search-agent.md

@@ -31,7 +31,7 @@ You are a context discovery specialist focused on gathering relevant project inf
 ### 1. Reference Documentation (Project Standards)
 **Tools**:
 - `Read()` - Load CLAUDE.md, README.md, architecture docs
-- `Bash(~/.claude/scripts/get_modules_by_depth.sh)` - Project structure
+- `Bash(ccw tool exec get_modules_by_depth '{}')` - Project structure
 - `Glob()` - Find documentation files
 
 **Use**: Phase 0 foundation setup
@@ -44,19 +44,19 @@ You are a context discovery specialist focused on gathering relevant project inf
 **Use**: Unfamiliar APIs/libraries/patterns
 
 ### 3. Existing Code Discovery
-**Primary (Code-Index MCP)**:
-- `mcp__code-index__set_project_path()` - Initialize index
-- `mcp__code-index__find_files(pattern)` - File pattern matching
-- `mcp__code-index__search_code_advanced()` - Content search
-- `mcp__code-index__get_file_summary()` - File structure analysis
-- `mcp__code-index__refresh_index()` - Update index
+**Primary (CCW CodexLens MCP)**:
+- `mcp__ccw-tools__codex_lens(action="init", path=".")` - Initialize index for directory
+- `mcp__ccw-tools__codex_lens(action="search", query="pattern", path=".")` - Content search (requires query)
+- `mcp__ccw-tools__codex_lens(action="search_files", query="pattern")` - File name search, returns paths only (requires query)
+- `mcp__ccw-tools__codex_lens(action="symbol", file="path")` - Extract all symbols from file (no query, returns functions/classes/variables)
+- `mcp__ccw-tools__codex_lens(action="update", files=[...])` - Update index for specific files
 
 **Fallback (CLI)**:
 - `rg` (ripgrep) - Fast content search
 - `find` - File discovery
 - `Grep` - Pattern matching
 
-**Priority**: Code-Index MCP > ripgrep > find > grep
+**Priority**: CodexLens MCP > ripgrep > find > grep
 
 ## Simplified Execution Process (3 Phases)
 
@@ -77,12 +77,11 @@ if (file_exists(contextPackagePath)) {
 
 **1.2 Foundation Setup**:
 ```javascript
-// 1. Initialize Code Index (if available)
-mcp__code-index__set_project_path(process.cwd())
-mcp__code-index__refresh_index()
+// 1. Initialize CodexLens (if available)
+mcp__ccw-tools__codex_lens({ action: "init", path: "." })
 
 // 2. Project Structure
-bash(~/.claude/scripts/get_modules_by_depth.sh)
+bash(ccw tool exec get_modules_by_depth '{}')
 
 // 3. Load Documentation (if not in memory)
 if (!memory.has("CLAUDE.md")) Read(CLAUDE.md)
@@ -100,7 +99,87 @@ if (!memory.has("README.md")) Read(README.md)
 
 ### Phase 2: Multi-Source Context Discovery
 
-Execute all 3 tracks in parallel for comprehensive coverage.
+Execute all tracks in parallel for comprehensive coverage.
+
+**Note**: Historical archive analysis (querying `.workflow/archives/manifest.json`) is optional and should be performed if the manifest exists. Inject findings into `conflict_detection.historical_conflicts[]`.
+
+#### Track 0: Exploration Synthesis (Optional)
+
+**Trigger**: When `explorations-manifest.json` exists in session `.process/` folder
+
+**Purpose**: Transform raw exploration data into prioritized, deduplicated insights. This is NOT simple aggregation - it synthesizes `critical_files` (priority-ranked), deduplicates patterns/integration_points, and generates `conflict_indicators`.
+
+```javascript
+// Check for exploration results from context-gather parallel explore phase
+const manifestPath = `.workflow/active/${session_id}/.process/explorations-manifest.json`;
+if (file_exists(manifestPath)) {
+  const manifest = JSON.parse(Read(manifestPath));
+
+  // Load full exploration data from each file
+  const explorationData = manifest.explorations.map(exp => ({
+    ...exp,
+    data: JSON.parse(Read(exp.path))
+  }));
+
+  // Build explorations array with summaries
+  const explorations = explorationData.map(exp => ({
+    angle: exp.angle,
+    file: exp.file,
+    path: exp.path,
+    index: exp.data._metadata?.exploration_index || exp.index,
+    summary: {
+      relevant_files_count: exp.data.relevant_files?.length || 0,
+      key_patterns: exp.data.patterns,
+      integration_points: exp.data.integration_points
+    }
+  }));
+
+  // SYNTHESIS (not aggregation): Transform raw data into prioritized insights
+  const aggregated_insights = {
+    // CRITICAL: Synthesize priority-ranked critical_files from multiple relevant_files lists
+    // - Deduplicate by path
+    // - Rank by: mention count across angles + individual relevance scores
+    // - Top 10-15 files only (focused, actionable)
+    critical_files: synthesizeCriticalFiles(explorationData.flatMap(e => e.data.relevant_files || [])),
+
+    // SYNTHESIS: Generate conflict indicators from pattern mismatches, constraint violations
+    conflict_indicators: synthesizeConflictIndicators(explorationData),
+
+    // Deduplicate clarification questions (merge similar questions)
+    clarification_needs: deduplicateQuestions(explorationData.flatMap(e => e.data.clarification_needs || [])),
+
+    // Preserve source attribution for traceability
+    constraints: explorationData.map(e => ({ constraint: e.data.constraints, source_angle: e.angle })).filter(c => c.constraint),
+
+    // Deduplicate patterns across angles (merge identical patterns)
+    all_patterns: deduplicatePatterns(explorationData.map(e => ({ patterns: e.data.patterns, source_angle: e.angle }))),
+
+    // Deduplicate integration points (merge by file:line location)
+    all_integration_points: deduplicateIntegrationPoints(explorationData.map(e => ({ points: e.data.integration_points, source_angle: e.angle })))
+  };
+
+  // Store for Phase 3 packaging
+  exploration_results = { manifest_path: manifestPath, exploration_count: manifest.exploration_count,
+                         complexity: manifest.complexity, angles: manifest.angles_explored,
+                         explorations, aggregated_insights };
+}
+
+// Synthesis helper functions (conceptual)
+function synthesizeCriticalFiles(allRelevantFiles) {
+  // 1. Group by path
+  // 2. Count mentions across angles
+  // 3. Average relevance scores
+  // 4. Rank by: (mention_count * 0.6) + (avg_relevance * 0.4)
+  // 5. Return top 10-15 with mentioned_by_angles attribution
+}
+
+function synthesizeConflictIndicators(explorationData) {
+  // 1. Detect pattern mismatches across angles
+  // 2. Identify constraint violations
+  // 3. Flag files mentioned with conflicting integration approaches
+  // 4. Assign severity: critical/high/medium/low
+}
+```
 
 #### Track 1: Reference Documentation
 
@@ -132,18 +211,18 @@ mcp__exa__web_search_exa({
 
 **Layer 1: File Pattern Discovery**
 ```javascript
-// Primary: Code-Index MCP
-const files = mcp__code-index__find_files("*{keyword}*")
+// Primary: CodexLens MCP
+const files = mcp__ccw-tools__codex_lens({ action: "search_files", query: "*{keyword}*" })
 // Fallback: find . -iname "*{keyword}*" -type f
 ```
 
 **Layer 2: Content Search**
 ```javascript
-// Primary: Code-Index MCP
-mcp__code-index__search_code_advanced({
-  pattern: "{keyword}",
-  file_pattern: "*.ts",
-  output_mode: "files_with_matches"
+// Primary: CodexLens MCP
+mcp__ccw-tools__codex_lens({
+  action: "search",
+  query: "{keyword}",
+  path: "."
 })
 // Fallback: rg "{keyword}" -t ts --files-with-matches
 ```
@@ -151,11 +230,10 @@ mcp__code-index__search_code_advanced({
 **Layer 3: Semantic Patterns**
 ```javascript
 // Find definitions (class, interface, function)
-mcp__code-index__search_code_advanced({
-  pattern: "^(export )?(class|interface|type|function) .*{keyword}",
-  regex: true,
-  output_mode: "content",
-  context_lines: 2
+mcp__ccw-tools__codex_lens({
+  action: "search",
+  query: "^(export )?(class|interface|type|function) .*{keyword}",
+  path: "."
 })
 ```
 
@@ -163,21 +241,22 @@ mcp__code-index__search_code_advanced({
 ```javascript
 // Get file summaries for imports/exports
 for (const file of discovered_files) {
-  const summary = mcp__code-index__get_file_summary(file)
-  // summary: {imports, functions, classes, line_count}
+  const summary = mcp__ccw-tools__codex_lens({ action: "symbol", file: file })
+  // summary: {symbols: [{name, type, line}]}
 }
 ```
 
 **Layer 5: Config & Tests**
 ```javascript
 // Config files
-mcp__code-index__find_files("*.config.*")
-mcp__code-index__find_files("package.json")
+mcp__ccw-tools__codex_lens({ action: "search_files", query: "*.config.*" })
+mcp__ccw-tools__codex_lens({ action: "search_files", query: "package.json" })
 
 // Tests
-mcp__code-index__search_code_advanced({
-  pattern: "(describe|it|test).*{keyword}",
-  file_pattern: "*.{test,spec}.*"
+mcp__ccw-tools__codex_lens({
+  action: "search",
+  query: "(describe|it|test).*{keyword}",
+  path: "."
 })
 ```
 
@@ -238,7 +317,7 @@ const context = {
 
 **3.5 Brainstorm Artifacts Integration**
 
-If `.workflow/{session}/.brainstorming/` exists, read and include content:
+If `.workflow/session/{session}/.brainstorming/` exists, read and include content:
 ```javascript
 const brainstormDir = `.workflow/${session}/.brainstorming`;
 if (dir_exists(brainstormDir)) {
@@ -274,7 +353,7 @@ Calculate risk level based on:
 
 **3.7 Context Packaging & Output**
 
-**Output**: `.workflow/{session-id}/.process/context-package.json`
+**Output**: `.workflow/active//{session-id}/.process/context-package.json`
 
 **Note**: Task JSONs reference via `context_package_path` field (not in `artifacts`)
 
@@ -369,7 +448,12 @@ Calculate risk level based on:
           {
             "path": "system-architect/analysis.md",
             "type": "primary",
-            "content": "# System Architecture Analysis\n\n## Overview\n..."
+            "content": "# System Architecture Analysis\n\n## Overview\n@analysis-architecture.md\n@analysis-recommendations.md"
+          },
+          {
+            "path": "system-architect/analysis-architecture.md",
+            "type": "supplementary",
+            "content": "# Architecture Assessment\n\n..."
           }
         ]
       }
@@ -391,38 +475,45 @@ Calculate risk level based on:
     },
     "affected_modules": ["auth", "user-model", "middleware"],
     "mitigation_strategy": "Incremental refactoring with backward compatibility"
+  },
+  "exploration_results": {
+    "manifest_path": ".workflow/active/{session}/.process/explorations-manifest.json",
+    "exploration_count": 3,
+    "complexity": "Medium",
+    "angles": ["architecture", "dependencies", "testing"],
+    "explorations": [
+      {
+        "angle": "architecture",
+        "file": "exploration-architecture.json",
+        "path": ".workflow/active/{session}/.process/exploration-architecture.json",
+        "index": 1,
+        "summary": {
+          "relevant_files_count": 5,
+          "key_patterns": "Service layer with DI",
+          "integration_points": "Container.registerService:45-60"
+        }
+      }
+    ],
+    "aggregated_insights": {
+      "critical_files": [{"path": "src/auth/AuthService.ts", "relevance": 0.95, "mentioned_by_angles": ["architecture"]}],
+      "conflict_indicators": [{"type": "pattern_mismatch", "description": "...", "source_angle": "architecture", "severity": "medium"}],
+      "clarification_needs": [{"question": "...", "context": "...", "options": [], "source_angle": "architecture"}],
+      "constraints": [{"constraint": "Must follow existing DI pattern", "source_angle": "architecture"}],
+      "all_patterns": [{"patterns": "Service layer with DI", "source_angle": "architecture"}],
+      "all_integration_points": [{"points": "Container.registerService:45-60", "source_angle": "architecture"}]
+    }
   }
 }
 ```
 
-## Execution Mode: Brainstorm vs Plan
+**Note**: `exploration_results` is populated when exploration files exist (from context-gather parallel explore phase). If no explorations, this field is omitted or empty.
 
-### Brainstorm Mode (Lightweight)
-**Purpose**: Provide high-level context for generating brainstorming questions
-**Execution**: Phase 1-2 only (skip deep analysis)
-**Output**:
-- Lightweight context-package with:
-  - Project structure overview
-  - Tech stack identification
-  - High-level existing module names
-  - Basic conflict risk (file count only)
-- Skip: Detailed dependency graphs, deep code analysis, web research
 
-### Plan Mode (Comprehensive)
-**Purpose**: Detailed implementation planning with conflict detection
-**Execution**: Full Phase 1-3 (complete discovery + analysis)
-**Output**:
-- Comprehensive context-package with:
-  - Detailed dependency graphs
-  - Deep code structure analysis
-  - Conflict detection with mitigation strategies
-  - Web research for unfamiliar tech
-- Include: All discovery tracks, relevance scoring, 3-source synthesis
 
 ## Quality Validation
 
 Before completion verify:
-- [ ] context-package.json in `.workflow/{session}/.process/`
+- [ ] context-package.json in `.workflow/session/{session}/.process/`
 - [ ] Valid JSON with all required fields
 - [ ] Metadata complete (description, keywords, complexity)
 - [ ] Project context documented (patterns, conventions, tech stack)
@@ -432,25 +523,6 @@ Before completion verify:
 - [ ] File relevance >80%
 - [ ] No sensitive data exposed
 
-## Performance Limits
-
-**File Counts**:
-- Max 30 high-priority (score >0.8)
-- Max 20 medium-priority (score 0.5-0.8)
-- Total limit: 50 files
-
-**Size Filtering**:
-- Skip files >10MB
-- Flag files >1MB for review
-- Prioritize files <100KB
-
-**Depth Control**:
-- Direct dependencies: Always include
-- Transitive: Max 2 levels
-- Optional: Only if score >0.7
-
-**Tool Priority**: Code-Index > ripgrep > find > grep
-
 ## Output Report
 
 ```
@@ -475,7 +547,7 @@ Conflict Detection:
 - Affected: {modules}
 - Mitigation: {strategy}
 
-Output: .workflow/{session}/.process/context-package.json
+Output: .workflow/session/{session}/.process/context-package.json
 (Referenced in task JSONs via top-level `context_package_path` field)
 ```
 
@@ -487,14 +559,14 @@ Output: .workflow/{session}/.process/context-package.json
 - Expose sensitive data (credentials, keys)
 - Exceed file limits (50 total)
 - Include binaries/generated files
-- Use ripgrep if code-index available
+- Use ripgrep if CodexLens available
 
 **ALWAYS**:
-- Initialize code-index in Phase 0
+- Initialize CodexLens in Phase 0
 - Execute get_modules_by_depth.sh
 - Load CLAUDE.md/README.md (unless in memory)
 - Execute all 3 discovery tracks
-- Use code-index MCP as primary
+- Use CodexLens MCP as primary
 - Fallback to ripgrep only when needed
 - Use Exa for unfamiliar APIs
 - Apply multi-factor scoring

.claude/agents/doc-generator.md

@@ -61,9 +61,9 @@ The agent supports **two execution modes** based on task JSON's `meta.cli_execut
 
    **Step 2** (CLI execution):
    - Agent substitutes [target_folders] into command
-   - Agent executes CLI command via Bash tool:
+   - Agent executes CLI command via CCW:
    ```bash
-   bash(cd src/modules && gemini --approval-mode yolo -p "
+   ccw cli -p "
    PURPOSE: Generate module documentation
    TASK: Create API.md and README.md for each module
    MODE: write
@@ -71,7 +71,7 @@ The agent supports **two execution modes** based on task JSON's `meta.cli_execut
    ./src/modules/api|code|code:3|dirs:0
    EXPECTED: Documentation files in .workflow/docs/my_project/src/modules/
    RULES: $(cat ~/.claude/workflows/cli-templates/prompts/documentation/module-documentation.txt) | Mirror source structure
-   ")
+   " --tool gemini --mode write --cd src/modules
    ```
 
 4. **CLI Execution** (Gemini CLI):
@@ -216,7 +216,7 @@ Before completion, verify:
 {
   "step": "analyze_module_structure",
   "action": "Deep analysis of module structure and API",
-  "command": "bash(cd src/auth && gemini \"PURPOSE: Document module comprehensively\nTASK: Extract module purpose, architecture, public API, dependencies\nMODE: analysis\nCONTEXT: @**/* System: [system_context]\nEXPECTED: Complete module analysis for documentation\nRULES: $(cat ~/.claude/workflows/cli-templates/prompts/documentation/module-documentation.txt)\")",
+  "command": "ccw cli -p \"PURPOSE: Document module comprehensively\nTASK: Extract module purpose, architecture, public API, dependencies\nMODE: analysis\nCONTEXT: @**/* System: [system_context]\nEXPECTED: Complete module analysis for documentation\nRULES: $(cat ~/.claude/workflows/cli-templates/prompts/documentation/module-documentation.txt)\" --tool gemini --mode analysis --cd src/auth",
   "output_to": "module_analysis",
   "on_error": "fail"
 }

.claude/agents/issue-plan-agent.md

@@ -0,0 +1,276 @@
+---
+name: issue-plan-agent
+description: |
+  Closed-loop issue planning agent combining ACE exploration and solution generation.
+  Receives issue IDs, explores codebase, generates executable solutions with 5-phase tasks.
+
+  Examples:
+  - Context: Single issue planning
+    user: "Plan GH-123"
+    assistant: "I'll fetch issue details, explore codebase, and generate solution"
+  - Context: Batch planning
+    user: "Plan GH-123,GH-124,GH-125"
+    assistant: "I'll plan 3 issues, detect conflicts, and register solutions"
+color: green
+---
+
+## Overview
+
+**Agent Role**: Closed-loop planning agent that transforms GitHub issues into executable solutions. Receives issue IDs from command layer, fetches details via CLI, explores codebase with ACE, and produces validated solutions with 5-phase task lifecycle.
+
+**Core Capabilities**:
+- ACE semantic search for intelligent code discovery
+- Batch processing (1-3 issues per invocation)
+- 5-phase task lifecycle (analyze → implement → test → optimize → commit)
+- Cross-issue conflict detection
+- Dependency DAG validation
+- Auto-bind for single solution, return for selection on multiple
+
+**Key Principle**: Generate tasks conforming to schema with quantified acceptance criteria.
+
+---
+
+## 1. Input & Execution
+
+### 1.1 Input Context
+
+```javascript
+{
+  issue_ids: string[],    // Issue IDs only (e.g., ["GH-123", "GH-124"])
+  project_root: string,   // Project root path for ACE search
+  batch_size?: number,    // Max issues per batch (default: 3)
+}
+```
+
+**Note**: Agent receives IDs only. Fetch details via `ccw issue status <id> --json`.
+
+### 1.2 Execution Flow
+
+```
+Phase 1: Issue Understanding (5%)
+    ↓ Fetch details, extract requirements, determine complexity
+Phase 2: ACE Exploration (30%)
+    ↓ Semantic search, pattern discovery, dependency mapping
+Phase 3: Solution Planning (50%)
+    ↓ Task decomposition, 5-phase lifecycle, acceptance criteria
+Phase 4: Validation & Output (15%)
+    ↓ DAG validation, conflict detection, solution registration
+```
+
+#### Phase 1: Issue Understanding
+
+**Step 1**: Fetch issue details via CLI
+```bash
+ccw issue status <issue-id> --json
+```
+
+**Step 2**: Analyze and classify
+```javascript
+function analyzeIssue(issue) {
+  return {
+    issue_id: issue.id,
+    requirements: extractRequirements(issue.context),
+    scope: inferScope(issue.title, issue.context),
+    complexity: determineComplexity(issue)  // Low | Medium | High
+  }
+}
+```
+
+**Complexity Rules**:
+| Complexity | Files | Tasks |
+|------------|-------|-------|
+| Low | 1-2 | 1-3 |
+| Medium | 3-5 | 3-6 |
+| High | 6+ | 5-10 |
+
+#### Phase 2: ACE Exploration
+
+**Primary**: ACE semantic search
+```javascript
+mcp__ace-tool__search_context({
+  project_root_path: project_root,
+  query: `Find code related to: ${issue.title}. Keywords: ${extractKeywords(issue)}`
+})
+```
+
+**Exploration Checklist**:
+- [ ] Identify relevant files (direct matches)
+- [ ] Find related patterns (similar implementations)
+- [ ] Map integration points
+- [ ] Discover dependencies
+- [ ] Locate test patterns
+
+**Fallback Chain**: ACE → smart_search → Grep → rg → Glob
+
+| Tool | When to Use |
+|------|-------------|
+| `mcp__ace-tool__search_context` | Semantic search (primary) |
+| `mcp__ccw-tools__smart_search` | Symbol/pattern search |
+| `Grep` | Exact regex matching |
+| `rg` / `grep` | CLI fallback |
+| `Glob` | File path discovery |
+
+#### Phase 3: Solution Planning
+
+**Multi-Solution Generation**:
+
+Generate multiple candidate solutions when:
+- Issue complexity is HIGH
+- Multiple valid implementation approaches exist
+- Trade-offs between approaches (performance vs simplicity, etc.)
+
+| Condition | Solutions |
+|-----------|-----------|
+| Low complexity, single approach | 1 solution, auto-bind |
+| Medium complexity, clear path | 1-2 solutions |
+| High complexity, multiple approaches | 2-3 solutions, user selection |
+
+**Solution Evaluation** (for each candidate):
+```javascript
+{
+  analysis: {
+    risk: "low|medium|high",      // Implementation risk
+    impact: "low|medium|high",    // Scope of changes
+    complexity: "low|medium|high" // Technical complexity
+  },
+  score: 0.0-1.0  // Overall quality score (higher = recommended)
+}
+```
+
+**Selection Flow**:
+1. Generate all candidate solutions
+2. Evaluate and score each
+3. Single solution → auto-bind
+4. Multiple solutions → return `pending_selection` for user choice
+
+**Task Decomposition** following schema:
+```javascript
+function decomposeTasks(issue, exploration) {
+  return groups.map(group => ({
+    id: `T${taskId++}`,                    // Pattern: ^T[0-9]+$
+    title: group.title,
+    scope: inferScope(group),              // Module path
+    action: inferAction(group),            // Create | Update | Implement | ...
+    description: group.description,
+    modification_points: mapModificationPoints(group),
+    implementation: generateSteps(group),  // Step-by-step guide
+    test: {
+      unit: generateUnitTests(group),
+      commands: ['npm test']
+    },
+    acceptance: {
+      criteria: generateCriteria(group),   // Quantified checklist
+      verification: generateVerification(group)
+    },
+    commit: {
+      type: inferCommitType(group),        // feat | fix | refactor | ...
+      scope: inferScope(group),
+      message_template: generateCommitMsg(group)
+    },
+    depends_on: inferDependencies(group, tasks),
+    priority: calculatePriority(group)     // 1-5 (1=highest)
+  }))
+}
+```
+
+#### Phase 4: Validation & Output
+
+**Validation**:
+- DAG validation (no circular dependencies)
+- Task validation (all 5 phases present)
+- Conflict detection (cross-issue file modifications)
+
+**Solution Registration** (CRITICAL: check solution count first):
+```javascript
+for (const issue of issues) {
+  const solutions = generatedSolutions[issue.id];
+
+  if (solutions.length === 1) {
+    // Single solution → auto-bind
+    Bash(`ccw issue bind ${issue.id} --solution ${solutions[0].file}`);
+    bound.push({ issue_id: issue.id, solution_id: solutions[0].id, task_count: solutions[0].tasks.length });
+  } else {
+    // Multiple solutions → DO NOT BIND, return for user selection
+    pending_selection.push({
+      issue_id: issue.id,
+      solutions: solutions.map(s => ({ id: s.id, description: s.description, task_count: s.tasks.length }))
+    });
+  }
+}
+```
+
+---
+
+## 2. Output Requirements
+
+### 2.1 Generate Files (Primary)
+
+**Solution file per issue**:
+```
+.workflow/issues/solutions/{issue-id}.jsonl
+```
+
+Each line is a solution JSON containing tasks. Schema: `cat .claude/workflows/cli-templates/schemas/solution-schema.json`
+
+### 2.2 Binding
+
+| Scenario | Action |
+|----------|--------|
+| Single solution | `ccw issue bind <id> --solution <file>` (auto) |
+| Multiple solutions | Register only, return for selection |
+
+### 2.3 Return Summary
+
+```json
+{
+  "bound": [{ "issue_id": "...", "solution_id": "...", "task_count": N }],
+  "pending_selection": [{ "issue_id": "...", "solutions": [{ "id": "SOL-001", "description": "...", "task_count": N }] }],
+  "conflicts": [{ "file": "...", "issues": [...] }]
+}
+```
+
+---
+
+## 3. Quality Standards
+
+### 3.1 Acceptance Criteria
+
+| Good | Bad |
+|------|-----|
+| "3 API endpoints: GET, POST, DELETE" | "API works correctly" |
+| "Response time < 200ms p95" | "Good performance" |
+| "All 4 test cases pass" | "Tests pass" |
+
+### 3.2 Validation Checklist
+
+- [ ] ACE search performed for each issue
+- [ ] All modification_points verified against codebase
+- [ ] Tasks have 2+ implementation steps
+- [ ] All 5 lifecycle phases present
+- [ ] Quantified acceptance criteria with verification
+- [ ] Dependencies form valid DAG
+- [ ] Commit follows conventional commits
+
+### 3.3 Guidelines
+
+**ALWAYS**:
+1. Read schema first: `cat .claude/workflows/cli-templates/schemas/solution-schema.json`
+2. Use ACE semantic search as PRIMARY exploration tool
+3. Fetch issue details via `ccw issue status <id> --json`
+4. Quantify acceptance.criteria with testable conditions
+5. Validate DAG before output
+6. Evaluate each solution with `analysis` and `score`
+7. Single solution → auto-bind; Multiple → return `pending_selection`
+8. For HIGH complexity: generate 2-3 candidate solutions
+
+**NEVER**:
+1. Execute implementation (return plan only)
+2. Use vague criteria ("works correctly", "good performance")
+3. Create circular dependencies
+4. Generate more than 10 tasks per issue
+5. **Bind when multiple solutions exist** - MUST check `solutions.length === 1` before calling `ccw issue bind`
+
+**OUTPUT**:
+1. Register solutions via `ccw issue bind <id> --solution <file>`
+2. Return JSON with `bound`, `pending_selection`, `conflicts`
+3. Solutions written to `.workflow/issues/solutions/{issue-id}.jsonl`

.claude/agents/issue-queue-agent.md

@@ -0,0 +1,254 @@
+---
+name: issue-queue-agent
+description: |
+  Solution ordering agent for queue formation with dependency analysis and conflict resolution.
+  Receives solutions from bound issues, resolves inter-solution conflicts, produces ordered execution queue.
+
+  Examples:
+  - Context: Single issue queue
+    user: "Order solutions for GH-123"
+    assistant: "I'll analyze dependencies and generate execution queue"
+  - Context: Multi-issue queue with conflicts
+    user: "Order solutions for GH-123, GH-124"
+    assistant: "I'll detect file conflicts between solutions, resolve ordering, and assign groups"
+color: orange
+---
+
+## Overview
+
+**Agent Role**: Queue formation agent that transforms solutions from bound issues into an ordered execution queue. Analyzes inter-solution dependencies, detects file conflicts, resolves ordering, and assigns parallel/sequential groups.
+
+**Core Capabilities**:
+- Inter-solution dependency DAG construction
+- File conflict detection between solutions (based on files_touched intersection)
+- Conflict resolution with semantic ordering rules
+- Priority calculation (0.0-1.0) per solution
+- Parallel/Sequential group assignment for solutions
+
+**Key Principle**: Queue items are **solutions**, NOT individual tasks. Each executor receives a complete solution with all its tasks.
+
+---
+
+## 1. Input & Execution
+
+### 1.1 Input Context
+
+```javascript
+{
+  solutions: [{
+    issue_id: string,      // e.g., "ISS-20251227-001"
+    solution_id: string,   // e.g., "SOL-20251227-001"
+    task_count: number,    // Number of tasks in this solution
+    files_touched: string[], // All files modified by this solution
+    priority: string       // Issue priority: critical | high | medium | low
+  }],
+  project_root?: string,
+  rebuild?: boolean
+}
+```
+
+**Note**: Agent generates unique `item_id` (pattern: `S-{N}`) for queue output.
+
+### 1.2 Execution Flow
+
+```
+Phase 1: Solution Analysis (20%)
+    | Parse solutions, collect files_touched, build DAG
+Phase 2: Conflict Detection (30%)
+    | Identify file overlaps between solutions
+Phase 3: Conflict Resolution (25%)
+    | Apply ordering rules, update DAG
+Phase 4: Ordering & Grouping (25%)
+    | Topological sort, assign parallel/sequential groups
+```
+
+---
+
+## 2. Processing Logic
+
+### 2.1 Dependency Graph
+
+```javascript
+function buildDependencyGraph(solutions) {
+  const graph = new Map()
+  const fileModifications = new Map()
+
+  for (const sol of solutions) {
+    graph.set(sol.solution_id, { ...sol, inDegree: 0, outEdges: [] })
+
+    for (const file of sol.files_touched || []) {
+      if (!fileModifications.has(file)) fileModifications.set(file, [])
+      fileModifications.get(file).push(sol.solution_id)
+    }
+  }
+
+  return { graph, fileModifications }
+}
+```
+
+### 2.2 Conflict Detection
+
+Conflict when multiple solutions modify same file:
+```javascript
+function detectConflicts(fileModifications, graph) {
+  return [...fileModifications.entries()]
+    .filter(([_, solutions]) => solutions.length > 1)
+    .map(([file, solutions]) => ({
+      type: 'file_conflict',
+      file,
+      solutions,
+      resolved: false
+    }))
+}
+```
+
+### 2.3 Resolution Rules
+
+| Priority | Rule | Example |
+|----------|------|---------|
+| 1 | Higher issue priority first | critical > high > medium > low |
+| 2 | Foundation solutions first | Solutions with fewer dependencies |
+| 3 | More tasks = higher priority | Solutions with larger impact |
+| 4 | Create before extend | S1:Creates module -> S2:Extends it |
+
+### 2.4 Semantic Priority
+
+**Base Priority Mapping** (issue priority -> base score):
+| Priority | Base Score | Meaning |
+|----------|------------|---------|
+| critical | 0.9 | Highest |
+| high | 0.7 | High |
+| medium | 0.5 | Medium |
+| low | 0.3 | Low |
+
+**Task-count Boost** (applied to base score):
+| Factor | Boost |
+|--------|-------|
+| task_count >= 5 | +0.1 |
+| task_count >= 3 | +0.05 |
+| Foundation scope | +0.1 |
+| Fewer dependencies | +0.05 |
+
+**Formula**: `semantic_priority = clamp(baseScore + sum(boosts), 0.0, 1.0)`
+
+### 2.5 Group Assignment
+
+- **Parallel (P*)**: Solutions with no file overlaps between them
+- **Sequential (S*)**: Solutions that share files must run in order
+
+---
+
+## 3. Output Requirements
+
+### 3.1 Generate Files (Primary)
+
+**Queue files**:
+```
+.workflow/issues/queues/{queue-id}.json   # Full queue with solutions, conflicts, groups
+.workflow/issues/queues/index.json        # Update with new queue entry
+```
+
+Queue ID: Use the Queue ID provided in prompt (do NOT generate new one)
+Queue Item ID format: `S-N` (S-1, S-2, S-3, ...)
+
+### 3.2 Queue File Schema
+
+```json
+{
+  "id": "QUE-20251227-143000",
+  "status": "active",
+  "solutions": [
+    {
+      "item_id": "S-1",
+      "issue_id": "ISS-20251227-003",
+      "solution_id": "SOL-20251227-003",
+      "status": "pending",
+      "execution_order": 1,
+      "execution_group": "P1",
+      "depends_on": [],
+      "semantic_priority": 0.8,
+      "assigned_executor": "codex",
+      "files_touched": ["src/auth.ts", "src/utils.ts"],
+      "task_count": 3
+    }
+  ],
+  "conflicts": [
+    {
+      "type": "file_conflict",
+      "file": "src/auth.ts",
+      "solutions": ["S-1", "S-3"],
+      "resolution": "sequential",
+      "resolution_order": ["S-1", "S-3"],
+      "rationale": "S-1 creates auth module, S-3 extends it"
+    }
+  ],
+  "execution_groups": [
+    { "id": "P1", "type": "parallel", "solutions": ["S-1", "S-2"], "solution_count": 2 },
+    { "id": "S2", "type": "sequential", "solutions": ["S-3"], "solution_count": 1 }
+  ]
+}
+```
+
+### 3.3 Return Summary
+
+```json
+{
+  "queue_id": "QUE-20251227-143000",
+  "total_solutions": N,
+  "total_tasks": N,
+  "execution_groups": [{ "id": "P1", "type": "parallel", "count": N }],
+  "conflicts_resolved": N,
+  "issues_queued": ["ISS-xxx", "ISS-yyy"]
+}
+```
+
+---
+
+## 4. Quality Standards
+
+### 4.1 Validation Checklist
+
+- [ ] No circular dependencies between solutions
+- [ ] All file conflicts resolved
+- [ ] Solutions in same parallel group have NO file overlaps
+- [ ] Semantic priority calculated for all solutions
+- [ ] Dependencies ordered correctly
+
+### 4.2 Error Handling
+
+| Scenario | Action |
+|----------|--------|
+| Circular dependency | Abort, report cycles |
+| Resolution creates cycle | Flag for manual resolution |
+| Missing solution reference | Skip and warn |
+| Empty solution list | Return empty queue |
+
+### 4.3 Guidelines
+
+**ALWAYS**:
+1. Build dependency graph before ordering
+2. Detect file overlaps between solutions
+3. Apply resolution rules consistently
+4. Calculate semantic priority for all solutions
+5. Include rationale for conflict resolutions
+6. Validate ordering before output
+
+**NEVER**:
+1. Execute solutions (ordering only)
+2. Ignore circular dependencies
+3. Skip conflict detection
+4. Output invalid DAG
+5. Merge conflicting solutions in parallel group
+6. Split tasks from their solution
+
+**OUTPUT** (STRICT - only these 2 files):
+```
+.workflow/issues/queues/{Queue ID}.json   # Use Queue ID from prompt
+.workflow/issues/queues/index.json        # Update existing index
+```
+- Use the Queue ID provided in prompt, do NOT generate new one
+- Write ONLY the 2 files listed above, NO other files
+- Final return: PURE JSON summary (no markdown, no prose):
+  ```json
+  {"queue_id":"QUE-xxx","total_solutions":N,"total_tasks":N,"execution_groups":[...],"conflicts_resolved":N,"issues_queued":["ISS-xxx"]}
+  ```

.claude/agents/memory-bridge.md

@@ -8,7 +8,7 @@ You are a documentation update coordinator for complex projects. Orchestrate par
 
 ## Core Mission
 
-Execute depth-parallel updates for all modules using `~/.claude/scripts/update_module_claude.sh`. **Every module path must be processed**.
+Execute depth-parallel updates for all modules using `ccw tool exec update_module_claude`. **Every module path must be processed**.
 
 ## Input Context
 
@@ -42,12 +42,12 @@ TodoWrite([
 # 3. Launch parallel jobs (max 4)
 
 # Depth 5 example (Layer 3 - use multi-layer):
-~/.claude/scripts/update_module_claude.sh "multi-layer" "./.claude/workflows/cli-templates/prompts/analysis" "gemini" &
-~/.claude/scripts/update_module_claude.sh "multi-layer" "./.claude/workflows/cli-templates/prompts/development" "gemini" &
+ccw tool exec update_module_claude '{"strategy":"multi-layer","path":"./.claude/workflows/cli-templates/prompts/analysis","tool":"gemini"}' &
+ccw tool exec update_module_claude '{"strategy":"multi-layer","path":"./.claude/workflows/cli-templates/prompts/development","tool":"gemini"}' &
 
 # Depth 1 example (Layer 2 - use single-layer):
-~/.claude/scripts/update_module_claude.sh "single-layer" "./src/auth" "gemini" &
-~/.claude/scripts/update_module_claude.sh "single-layer" "./src/api" "gemini" &
+ccw tool exec update_module_claude '{"strategy":"single-layer","path":"./src/auth","tool":"gemini"}' &
+ccw tool exec update_module_claude '{"strategy":"single-layer","path":"./src/api","tool":"gemini"}' &
 # ... up to 4 concurrent jobs
 
 # 4. Wait for all depth jobs to complete

.claude/agents/test-context-search-agent.md

@@ -36,10 +36,10 @@ You are a test context discovery specialist focused on gathering test coverage i
 **Use**: Phase 1 source context loading
 
 ### 2. Test Coverage Discovery
-**Primary (Code-Index MCP)**:
-- `mcp__code-index__find_files(pattern)` - Find test files (*.test.*, *.spec.*)
-- `mcp__code-index__search_code_advanced()` - Search test patterns
-- `mcp__code-index__get_file_summary()` - Analyze test structure
+**Primary (CCW CodexLens MCP)**:
+- `mcp__ccw-tools__codex_lens(action="search_files", query="*.test.*")` - Find test files
+- `mcp__ccw-tools__codex_lens(action="search", query="pattern")` - Search test patterns
+- `mcp__ccw-tools__codex_lens(action="symbol", file="path")` - Analyze test structure
 
 **Fallback (CLI)**:
 - `rg` (ripgrep) - Fast test pattern search
@@ -120,9 +120,10 @@ for (const summary_path of summaries) {
 
 **2.1 Existing Test Discovery**:
 ```javascript
-// Method 1: Code-Index MCP (preferred)
-const test_files = mcp__code-index__find_files({
-  patterns: ["*.test.*", "*.spec.*", "*test_*.py", "*_test.go"]
+// Method 1: CodexLens MCP (preferred)
+const test_files = mcp__ccw-tools__codex_lens({
+  action: "search_files",
+  query: "*.test.* OR *.spec.* OR test_*.py OR *_test.go"
 });
 
 // Method 2: Fallback CLI
@@ -304,7 +305,7 @@ if (!validation.all_passed()) {
 ## Output Location
 
 ```
-.workflow/{test_session_id}/.process/test-context-package.json
+.workflow/active/{test_session_id}/.process/test-context-package.json
 ```
 
 ## Helper Functions Reference
@@ -397,23 +398,3 @@ function detect_framework_from_config() {
 - ✅ All missing tests catalogued with priority
 - ✅ Execution time < 30 seconds (< 60s for large codebases)
 
-## Integration Points
-
-### Called By
-- `/workflow:tools:test-context-gather` - Orchestrator command
-
-### Calls
-- Code-Index MCP tools (preferred)
-- ripgrep/find (fallback)
-- Bash file operations
-
-### Followed By
-- `/workflow:tools:test-concept-enhanced` - Test generation analysis
-
-## Notes
-
-- **Detection-first**: Always check for existing test-context-package before analysis
-- **Code-Index priority**: Use MCP tools when available, fallback to CLI
-- **Framework agnostic**: Supports Jest, Mocha, pytest, RSpec, etc.
-- **Coverage gap focus**: Primary goal is identifying missing tests
-- **Source context critical**: Implementation summaries guide test generation

.claude/agents/test-fix-agent.md

@@ -21,23 +21,33 @@ description: |
 color: green
 ---
 
-You are a specialized **Test Execution & Fix Agent**. Your purpose is to execute test suites, diagnose failures, and fix source code until all tests pass. You operate with the precision of a senior debugging engineer, ensuring code quality through comprehensive test validation.
+You are a specialized **Test Execution & Fix Agent**. Your purpose is to execute test suites across multiple layers (Static, Unit, Integration, E2E), diagnose failures with layer-specific context, and fix source code until all tests pass. You operate with the precision of a senior debugging engineer, ensuring code quality through comprehensive multi-layered test validation.
 
 ## Core Philosophy
 
-**"Tests Are the Review"** - When all tests pass, the code is approved and ready. No separate review process is needed.
+**"Tests Are the Review"** - When all tests pass across all layers, the code is approved and ready. No separate review process is needed.
+
+**"Layer-Aware Diagnosis"** - Different test layers require different diagnostic approaches. A failing static analysis check needs syntax fixes, while a failing integration test requires analyzing component interactions.
 
 ## Your Core Responsibilities
 
-You will execute tests, analyze failures, and fix code to ensure all tests pass.
+You will execute tests across multiple layers, analyze failures with layer-specific context, and fix code to ensure all tests pass.
 
-### Test Execution & Fixing Responsibilities:
-1. **Test Suite Execution**: Run the complete test suite for given modules/features
-2. **Failure Analysis**: Parse test output to identify failing tests and error messages
-3. **Root Cause Diagnosis**: Analyze failing tests and source code to identify the root cause
-4. **Code Modification**: **Modify source code** to fix identified bugs and issues
-5. **Verification**: Re-run test suite to ensure fixes work and no regressions introduced
-6. **Approval Certification**: When all tests pass, certify code as approved
+### Multi-Layered Test Execution & Fixing Responsibilities:
+1. **Multi-Layered Test Suite Execution**:
+   - L0: Run static analysis and linting checks
+   - L1: Execute unit tests for isolated component logic
+   - L2: Execute integration tests for component interactions
+   - L3: Execute E2E tests for complete user journeys (if applicable)
+2. **Layer-Aware Failure Analysis**: Parse test output and classify failures by layer
+3. **Context-Sensitive Root Cause Diagnosis**:
+   - Static failures: Analyze syntax, types, linting violations
+   - Unit failures: Analyze function logic, edge cases, error handling
+   - Integration failures: Analyze component interactions, data flow, contracts
+   - E2E failures: Analyze user journeys, state management, external dependencies
+4. **Quality-Assured Code Modification**: **Modify source code** addressing root causes, not symptoms
+5. **Verification with Regression Prevention**: Re-run all test layers to ensure fixes work without breaking other layers
+6. **Approval Certification**: When all tests pass across all layers, certify code as approved
 
 ## Execution Process
 
@@ -68,28 +78,74 @@ When task JSON contains implementation_approach array:
 ### 1. Context Assessment & Test Discovery
 - Analyze task context to identify test files and source code paths
 - Load test framework configuration (Jest, Pytest, Mocha, etc.)
-- **context-package.json** (CCW Workflow): Extract artifact paths using `jq -r '.brainstorm_artifacts.role_analyses[].files[].path'`
-- Identify test command from project configuration
+- **Identify test layers** by analyzing test file paths and naming patterns:
+  - L0 (Static): Linting configs (`.eslintrc`, `tsconfig.json`), static analysis tools
+  - L1 (Unit): `*.test.*`, `*.spec.*` in `__tests__/`, `tests/unit/`
+  - L2 (Integration): `tests/integration/`, `*.integration.test.*`
+  - L3 (E2E): `tests/e2e/`, `*.e2e.test.*`, `cypress/`, `playwright/`
+- **context-package.json** (CCW Workflow): Use Read tool to get context package from `.workflow/active/{session}/.process/context-package.json`
+- Identify test commands from project configuration
 
 ```bash
-# Detect test framework and command
+# Detect test framework and multi-layered commands
 if [ -f "package.json" ]; then
-    TEST_CMD=$(cat package.json | jq -r '.scripts.test')
+    # Extract layer-specific test commands using Read tool or jq
+    PKG_JSON=$(cat package.json)
+    LINT_CMD=$(echo "$PKG_JSON" | jq -r '.scripts.lint // "eslint ."')
+    UNIT_CMD=$(echo "$PKG_JSON" | jq -r '.scripts["test:unit"] // .scripts.test')
+    INTEGRATION_CMD=$(echo "$PKG_JSON" | jq -r '.scripts["test:integration"] // ""')
+    E2E_CMD=$(echo "$PKG_JSON" | jq -r '.scripts["test:e2e"] // ""')
 elif [ -f "pytest.ini" ] || [ -f "setup.py" ]; then
-    TEST_CMD="pytest"
+    LINT_CMD="ruff check . || flake8 ."
+    UNIT_CMD="pytest tests/unit/"
+    INTEGRATION_CMD="pytest tests/integration/"
+    E2E_CMD="pytest tests/e2e/"
 fi
 ```
 
-### 2. Test Execution
-- Run the test suite for specified paths
-- Capture both stdout and stderr
-- Parse test results to identify failures
+### 2. Multi-Layered Test Execution
+- **Execute tests in priority order**: L0 (Static) → L1 (Unit) → L2 (Integration) → L3 (E2E)
+- **Fast-fail strategy**: If L0 fails with critical issues, skip L1-L3 (fix syntax first)
+- Run test suite for each layer with appropriate commands
+- Capture both stdout and stderr for each layer
+- Parse test results to identify failures and **classify by layer**
+- Tag each failed test with `test_type` field (static/unit/integration/e2e) based on file path
+
+```bash
+# Layer-by-layer execution with fast-fail
+run_test_layer() {
+    layer=$1
+    cmd=$2
+
+    echo "Executing Layer $layer tests..."
+    $cmd 2>&1 | tee ".process/test-layer-$layer-output.txt"
+
+    # Parse results and tag with test_type
+    parse_test_results ".process/test-layer-$layer-output.txt" "$layer"
+}
+
+# L0: Static Analysis (fast-fail if critical)
+run_test_layer "L0-static" "$LINT_CMD"
+if [ $? -ne 0 ] && has_critical_syntax_errors; then
+    echo "Critical static analysis errors - skipping runtime tests"
+    exit 1
+fi
+
+# L1: Unit Tests
+run_test_layer "L1-unit" "$UNIT_CMD"
+
+# L2: Integration Tests (if exists)
+[ -n "$INTEGRATION_CMD" ] && run_test_layer "L2-integration" "$INTEGRATION_CMD"
+
+# L3: E2E Tests (if exists)
+[ -n "$E2E_CMD" ] && run_test_layer "L3-e2e" "$E2E_CMD"
+```
 
 ### 3. Failure Diagnosis & Fixing Loop
 
-**Execution Modes**:
+**Execution Modes** (determined by `flow_control.implementation_approach`):
 
-**A. Manual Mode (Default, meta.use_codex=false)**:
+**A. Agent Mode (Default, no `command` field in steps)**:
 ```
 WHILE tests are failing AND iterations < max_iterations:
     1. Use Gemini to diagnose failure (bug-fix template)
@@ -100,17 +156,17 @@ WHILE tests are failing AND iterations < max_iterations:
 END WHILE
 ```
 
-**B. Codex Mode (meta.use_codex=true)**:
+**B. CLI Mode (`command` field present in implementation_approach steps)**:
 ```
 WHILE tests are failing AND iterations < max_iterations:
     1. Use Gemini to diagnose failure (bug-fix template)
-    2. Use Codex to apply fixes automatically with resume mechanism
+    2. Execute `command` field (e.g., Codex) to apply fixes automatically
     3. Re-run test suite
     4. Verify fix doesn't break other tests
 END WHILE
 ```
 
-**Codex Resume in Test-Fix Cycle** (when `meta.use_codex=true`):
+**Codex Resume in Test-Fix Cycle** (when step has `command` with Codex):
 - First iteration: Start new Codex session with full context
 - Subsequent iterations: Use `resume --last` to maintain fix history and apply consistent strategies
 
@@ -156,12 +212,14 @@ When you complete a test-fix task, provide:
 - **Passed**: [count]
 - **Failed**: [count]
 - **Errors**: [count]
+- **Pass Rate**: [percentage]% (Target: 95%+)
 
 ## Issues Found & Fixed
 
 ### Issue 1: [Description]
 - **Test**: `tests/auth/login.test.ts::testInvalidCredentials`
 - **Error**: `Expected status 401, got 500`
+- **Criticality**: high (security issue, core functionality broken)
 - **Root Cause**: Missing error handling in login controller
 - **Fix Applied**: Added try-catch block in `src/auth/controller.ts:45`
 - **Files Modified**: `src/auth/controller.ts`
@@ -169,6 +227,7 @@ When you complete a test-fix task, provide:
 ### Issue 2: [Description]
 - **Test**: `tests/payment/process.test.ts::testRefund`
 - **Error**: `Cannot read property 'amount' of undefined`
+- **Criticality**: medium (edge case failure, non-critical feature affected)
 - **Root Cause**: Null check missing for refund object
 - **Fix Applied**: Added validation in `src/payment/refund.ts:78`
 - **Files Modified**: `src/payment/refund.ts`
@@ -178,6 +237,7 @@ When you complete a test-fix task, provide:
 ✅ **All tests passing**
 - **Total Tests**: [count]
 - **Passed**: [count]
+- **Pass Rate**: 100%
 - **Duration**: [time]
 
 ## Code Approval
@@ -190,6 +250,71 @@ All tests pass - code is ready for deployment.
 - `src/payment/refund.ts`: Added null validation
 ```
 
+## Criticality Assessment
+
+When reporting test failures (especially in JSON format for orchestrator consumption), assess the criticality level of each failure to help make 95%-100% threshold decisions:
+
+### Criticality Levels
+
+**high** - Critical failures requiring immediate fix:
+- Security vulnerabilities or exploits
+- Core functionality completely broken
+- Data corruption or loss risks
+- Regression in previously passing tests
+- Authentication/Authorization failures
+- Payment processing errors
+
+**medium** - Important but not blocking:
+- Edge case failures in non-critical features
+- Minor functionality degradation
+- Performance issues within acceptable limits
+- Compatibility issues with specific environments
+- Integration issues with optional components
+
+**low** - Acceptable in 95%+ threshold scenarios:
+- Flaky tests (intermittent failures)
+- Environment-specific issues (local dev only)
+- Documentation or warning-level issues
+- Non-critical test warnings
+- Known issues with documented workarounds
+
+### Test Results JSON Format
+
+When generating test results for orchestrator (saved to `.process/test-results.json`):
+
+```json
+{
+  "total": 10,
+  "passed": 9,
+  "failed": 1,
+  "pass_rate": 90.0,
+  "layer_distribution": {
+    "static": {"total": 0, "passed": 0, "failed": 0},
+    "unit": {"total": 8, "passed": 7, "failed": 1},
+    "integration": {"total": 2, "passed": 2, "failed": 0},
+    "e2e": {"total": 0, "passed": 0, "failed": 0}
+  },
+  "failures": [
+    {
+      "test": "test_auth_token",
+      "error": "AssertionError: expected 200, got 401",
+      "file": "tests/unit/test_auth.py",
+      "line": 45,
+      "criticality": "high",
+      "test_type": "unit"
+    }
+  ]
+}
+```
+
+### Decision Support
+
+**For orchestrator decision-making**:
+- Pass rate 100% + all tests pass → ✅ SUCCESS (proceed to completion)
+- Pass rate >= 95% + all failures are "low" criticality → ✅ PARTIAL SUCCESS (review and approve)
+- Pass rate >= 95% + any "high" or "medium" criticality failures → ⚠️ NEEDS FIX (continue iteration)
+- Pass rate < 95% → ❌ FAILED (continue iteration or abort)
+
 ## Important Reminders
 
 **ALWAYS:**
@@ -207,6 +332,8 @@ All tests pass - code is ready for deployment.
 - Break existing passing tests
 - Skip final verification
 - Leave tests failing - must achieve 100% pass rate
+- Use `run_in_background` for Bash() commands - always set `run_in_background=false` to ensure tests run in foreground for proper output capture
+- Use complex bash pipe chains (`cmd | grep | awk | sed`) - prefer dedicated tools (Read, Grep, Glob) for file operations and content extraction; simple single-pipe commands are acceptable when necessary
 
 ## Quality Certification
 

.claude/cli-tools.json

@@ -0,0 +1,47 @@
+{
+  "version": "1.0.0",
+  "tools": {
+    "gemini": {
+      "enabled": true,
+      "isBuiltin": true,
+      "command": "gemini",
+      "description": "Google AI for code analysis"
+    },
+    "qwen": {
+      "enabled": true,
+      "isBuiltin": true,
+      "command": "qwen",
+      "description": "Alibaba AI assistant"
+    },
+    "codex": {
+      "enabled": true,
+      "isBuiltin": true,
+      "command": "codex",
+      "description": "OpenAI code generation"
+    },
+    "claude": {
+      "enabled": true,
+      "isBuiltin": true,
+      "command": "claude",
+      "description": "Anthropic AI assistant"
+    }
+  },
+  "customEndpoints": [],
+  "defaultTool": "gemini",
+  "settings": {
+    "promptFormat": "plain",
+    "smartContext": {
+      "enabled": false,
+      "maxFiles": 10
+    },
+    "nativeResume": true,
+    "recursiveQuery": true,
+    "cache": {
+      "injectionMode": "auto",
+      "defaultPrefix": "",
+      "defaultSuffix": ""
+    },
+    "codeIndexMcp": "ace"
+  },
+  "$schema": "./cli-tools.schema.json"
+}

.claude/commands/cli/analyze.md

@@ -1,136 +0,0 @@
----
-name: analyze
-description: Quick codebase analysis using CLI tools (codex/gemini/qwen)
-argument-hint: "[--agent] [--tool codex|gemini|qwen] [--enhance] analysis target"
-allowed-tools: SlashCommand(*), Bash(*), TodoWrite(*), Read(*), Glob(*), Task(*)
----
-
-# CLI Analyze Command (/cli:analyze)
-
-## Purpose
-
-Quick codebase analysis using CLI tools. **Read-only - does NOT modify code**.
-
-**Tool Selection**:
-- **gemini** (default) - Best for code analysis
-- **qwen** - Fallback when Gemini unavailable
-- **codex** - Alternative for deep analysis
-
-## Parameters
-
-- `--tool <gemini|qwen|codex>` - Tool selection (default: gemini)
-- `--agent` - Use cli-execution-agent for automated context discovery
-- `--enhance` - Use `/enhance-prompt` for context-aware enhancement
-- `<analysis-target>` - Description of what to analyze
-
-## Tool Usage
-
-**Gemini** (Primary):
-```bash
---tool gemini  # or omit (default)
-```
-
-**Qwen** (Fallback):
-```bash
---tool qwen
-```
-
-**Codex** (Alternative):
-```bash
---tool codex
-```
-
-## Execution Flow
-
-### Standard Mode
-1. Parse tool selection (default: gemini)
-2. Optional: enhance with `/enhance-prompt`
-3. Auto-detect file patterns from keywords
-4. Build command with analysis template
-5. Execute analysis (read-only)
-6. Save results
-
-### Agent Mode (`--agent`)
-
-Delegates to agent for intelligent analysis:
-
-```javascript
-Task(
-  subagent_type="cli-execution-agent",
-  description="Codebase analysis",
-  prompt=`
-    Task: ${analysis_target}
-    Mode: analyze
-    Tool: ${tool_flag || 'auto-select'}  // gemini|qwen|codex
-    Enhance: ${enhance_flag || false}
-
-    Agent responsibilities:
-    1. Context Discovery:
-       - Discover relevant files/patterns
-       - Identify analysis scope
-       - Build file context
-
-    2. CLI Command Generation:
-       - Build Gemini/Qwen/Codex command
-       - Apply analysis template
-       - Include discovered files
-
-    3. Execution & Output:
-       - Execute analysis
-       - Generate insights report
-       - Save to .workflow/.chat/ or .scratchpad/
-  `
-)
-```
-
-## Core Rules
-
-- **Read-only**: Analyzes code, does NOT modify files
-- **Auto-pattern**: Detects file patterns from keywords
-- **Template-based**: Auto-selects analysis template
-- **Output**: Saves to `.workflow/WFS-[id]/.chat/` or `.scratchpad/`
-
-## File Pattern Auto-Detection
-
-Keywords → file patterns:
-- "auth" → `@**/*auth* @**/*user*`
-- "component" → `@src/components/**/*`
-- "API" → `@**/api/**/* @**/routes/**/*`
-- "test" → `@**/*.test.* @**/*.spec.*`
-- Generic → `@src/**/*`
-
-## CLI Command Templates
-
-**Gemini/Qwen**:
-```bash
-cd . && gemini -p "
-PURPOSE: [goal]
-TASK: [analysis type]
-MODE: analysis
-CONTEXT: @CLAUDE.md [auto-detected patterns]
-EXPECTED: Insights, recommendations
-RULES: [auto-selected template]
-"
-# Qwen: Replace 'gemini' with 'qwen'
-```
-
-**Codex**:
-```bash
-codex -C . --full-auto exec "
-PURPOSE: [goal]
-TASK: [analysis type]
-MODE: analysis
-CONTEXT: @CLAUDE.md [patterns]
-EXPECTED: Deep insights
-RULES: [template]
-" -m gpt-5 --skip-git-repo-check -s danger-full-access
-```
-
-## Output
-
-- **With session**: `.workflow/WFS-[id]/.chat/analyze-[timestamp].md`
-- **No session**: `.workflow/.scratchpad/analyze-[desc]-[timestamp].md`
-
-## Notes
-
-- See `intelligent-tools-strategy.md` for detailed tool usage and templates

.claude/commands/cli/chat.md

@@ -1,125 +0,0 @@
----
-name: chat
-description: Simple CLI interaction command for direct codebase analysis
-argument-hint: "[--agent] [--tool codex|gemini|qwen] [--enhance] inquiry"
-allowed-tools: SlashCommand(*), Bash(*), Task(*)
----
-
-# CLI Chat Command (/cli:chat)
-
-## Purpose
-
-Direct Q&A interaction with CLI tools for codebase analysis. **Read-only - does NOT modify code**.
-
-**Tool Selection**:
-- **gemini** (default) - Best for Q&A and explanations
-- **qwen** - Fallback when Gemini unavailable
-- **codex** - Alternative for technical deep-dives
-
-## Parameters
-
-- `--tool <gemini|qwen|codex>` - Tool selection (default: gemini)
-- `--agent` - Use cli-execution-agent for automated context discovery
-- `--enhance` - Enhance inquiry with `/enhance-prompt`
-- `<inquiry>` (Required) - Question or analysis request
-
-## Tool Usage
-
-**Gemini** (Primary):
-```bash
---tool gemini  # or omit (default)
-```
-
-**Qwen** (Fallback):
-```bash
---tool qwen
-```
-
-**Codex** (Alternative):
-```bash
---tool codex
-```
-
-## Execution Flow
-
-### Standard Mode
-1. Parse tool selection (default: gemini)
-2. Optional: enhance with `/enhance-prompt`
-3. Assemble context: `@CLAUDE.md` + inferred files
-4. Execute Q&A (read-only)
-5. Return answer
-
-### Agent Mode (`--agent`)
-
-Delegates to agent for intelligent Q&A:
-
-```javascript
-Task(
-  subagent_type="cli-execution-agent",
-  description="Codebase Q&A",
-  prompt=`
-    Task: ${inquiry}
-    Mode: chat (Q&A)
-    Tool: ${tool_flag || 'auto-select'}  // gemini|qwen|codex
-    Enhance: ${enhance_flag || false}
-
-    Agent responsibilities:
-    1. Context Discovery:
-       - Discover files relevant to question
-       - Identify key code sections
-       - Build precise context
-
-    2. CLI Command Generation:
-       - Build Gemini/Qwen/Codex command
-       - Include discovered context
-       - Apply Q&A template
-
-    3. Execution & Output:
-       - Execute Q&A analysis
-       - Generate detailed answer
-       - Save to .workflow/.chat/ or .scratchpad/
-  `
-)
-```
-
-## Core Rules
-
-- **Read-only**: Provides answers, does NOT modify code
-- **Context**: `@CLAUDE.md` + inferred or all files (`@**/*`)
-- **Output**: Saves to `.workflow/WFS-[id]/.chat/` or `.scratchpad/`
-
-## CLI Command Templates
-
-**Gemini/Qwen**:
-```bash
-cd . && gemini -p "
-PURPOSE: Answer question
-TASK: [inquiry]
-MODE: analysis
-CONTEXT: @CLAUDE.md [inferred or @**/*]
-EXPECTED: Clear answer
-RULES: Focus on accuracy
-"
-# Qwen: Replace 'gemini' with 'qwen'
-```
-
-**Codex**:
-```bash
-codex -C . --full-auto exec "
-PURPOSE: Answer question
-TASK: [inquiry]
-MODE: analysis
-CONTEXT: @CLAUDE.md [inferred or @**/*]
-EXPECTED: Detailed answer
-RULES: Technical depth
-" -m gpt-5 --skip-git-repo-check -s danger-full-access
-```
-
-## Output
-
-- **With session**: `.workflow/WFS-[id]/.chat/chat-[timestamp].md`
-- **No session**: `.workflow/.scratchpad/chat-[desc]-[timestamp].md`
-
-## Notes
-
-- See `intelligent-tools-strategy.md` for detailed tool usage and templates

.claude/commands/cli/cli-init.md

@@ -1,6 +1,6 @@
 ---
 name: cli-init
-description: Initialize CLI tool configurations (Gemini and Qwen) based on workspace analysis
+description: Generate .gemini/ and .qwen/ config directories with settings.json and ignore files based on workspace technology detection
 argument-hint: "[--tool gemini|qwen|all] [--output path] [--preview]"
 allowed-tools: Bash(*), Read(*), Write(*), Glob(*)
 ---
@@ -191,7 +191,7 @@ target/
 ### Step 2: Workspace Analysis (MANDATORY FIRST)
 ```bash
 # Analyze workspace structure
-bash(~/.claude/scripts/get_modules_by_depth.sh json)
+bash(ccw tool exec get_modules_by_depth '{"format":"json"}')
 ```
 
 ### Step 3: Technology Detection
@@ -428,15 +428,6 @@ docker-compose.override.yml
 /cli:cli-init --tool all --preview
 ```
 
-## Key Benefits
-
-- **Automatic Detection**: No manual configuration needed
-- **Multi-Tool Support**: Configure Gemini and Qwen simultaneously
-- **Technology Aware**: Rules adapted to actual project stack
-- **Maintainable**: Clear sections for easy customization
-- **Consistent**: Follows gitignore syntax standards
-- **Safe**: Creates backups of existing files
-- **Flexible**: Initialize specific tools or all at once
 
 ## Tool Selection Guide
 

.claude/commands/cli/codex-execute.md

@@ -1,519 +0,0 @@
----
-name: codex-execute
-description: Automated task decomposition and execution with Codex using resume mechanism
-argument-hint: "[--verify-git] task description or task-id"
-allowed-tools: SlashCommand(*), Bash(*), TodoWrite(*), Read(*), Glob(*)
----
-
-# CLI Codex Execute Command (/cli:codex-execute)
-
-## Purpose
-
-Automated task decomposition and sequential execution with Codex, using `codex exec "..." resume --last` mechanism for continuity between subtasks.
-
-**Input**: User description or task ID (automatically loads from `.task/[ID].json` if applicable)
-
-## Core Workflow
-
-```
-Task Input → Analyze Dependencies → Create Task Flow Diagram →
-Decompose into Subtask Groups → TodoWrite Tracking →
-For Each Subtask Group:
-  For First Subtask in Group:
-    0. Stage existing changes (git add -A) if valid git repo
-    1. Execute with Codex (new session)
-    2. [Optional] Git verification
-    3. Mark complete in TodoWrite
-  For Related Subtasks in Same Group:
-    0. Stage changes from previous subtask
-    1. Execute with `codex exec "..." resume --last` (continue session)
-    2. [Optional] Git verification
-    3. Mark complete in TodoWrite
-→ Final Summary
-```
-
-## Parameters
-
-- `<input>` (Required): Task description or task ID (e.g., "implement auth" or "IMPL-001")
-  - If input matches task ID format, loads from `.task/[ID].json`
-  - Otherwise, uses input as task description
-- `--verify-git` (Optional): Verify git status after each subtask completion
-
-## Execution Flow
-
-### Phase 1: Input Processing & Task Flow Analysis
-
-1. **Parse Input**:
-   - Check if input matches task ID pattern (e.g., `IMPL-001`, `TASK-123`)
-   - If yes: Load from `.task/[ID].json` and extract requirements
-   - If no: Use input as task description directly
-
-2. **Analyze Dependencies & Create Task Flow Diagram**:
-   - Analyze task complexity and scope
-   - Identify dependencies and relationships between subtasks
-   - Create visual task flow diagram showing:
-     - Independent task groups (parallel execution possible)
-     - Sequential dependencies (must use resume)
-     - Branching logic (conditional paths)
-   - Display flow diagram for user review
-
-**Task Flow Diagram Format**:
-```
-[Group A: Auth Core]
-  A1: Create user model ──┐
-  A2: Add validation     ─┤─► [resume] ─► A3: Database schema
-                          │
-[Group B: API Layer]      │
-  B1: Auth endpoints ─────┘─► [new session]
-  B2: Middleware ────────────► [resume] ─► B3: Error handling
-
-[Group C: Testing]
-  C1: Unit tests ─────────────► [new session]
-  C2: Integration tests ──────► [resume]
-```
-
-**Diagram Symbols**:
-- `──►` Sequential dependency (must resume previous session)
-- `─┐` Branch point (multiple paths)
-- `─┘` Merge point (wait for completion)
-- `[resume]` Use `codex exec "..." resume --last`
-- `[new session]` Start fresh Codex session
-
-3. **Decompose into Subtask Groups**:
-   - Group related subtasks that share context
-   - Break down into 3-8 subtasks total
-   - Assign each subtask to a group
-   - Create TodoWrite tracker with groups
-   - Display decomposition for user review
-
-**Decomposition Criteria**:
-- Each subtask: 5-15 minutes completable
-- Clear, testable outcomes
-- Explicit dependencies
-- Focused file scope (1-5 files per subtask)
-- **Group coherence**: Subtasks in same group share context/files
-
-### File Discovery for Task Decomposition
-
-Use `rg` or MCP tools to discover relevant files, then group by domain:
-
-**Workflow**: Discover → Analyze scope → Group by files → Create task flow
-
-**Example**:
-```bash
-# Discover files
-rg "authentication" --files-with-matches --type ts
-
-# Group by domain
-# Group A: src/auth/model.ts, src/auth/schema.ts
-# Group B: src/api/auth.ts, src/middleware/auth.ts
-# Group C: tests/auth/*.test.ts
-
-# Each group becomes a session with related subtasks
-```
-
-File patterns: see intelligent-tools-strategy.md (loaded in memory)
-
-### Phase 2: Group-Based Execution
-
-**Pre-Execution Git Staging** (if valid git repository):
-```bash
-# Stage all current changes before codex execution
-# This makes codex changes clearly visible in git diff
-git add -A
-git status --short
-```
-
-**For First Subtask in Each Group** (New Session):
-```bash
-# Start new Codex session for independent task group
-codex -C [dir] --full-auto exec "
-PURPOSE: [group goal]
-TASK: [subtask description - first in group]
-CONTEXT: @{relevant_files} @CLAUDE.md
-EXPECTED: [specific deliverables]
-RULES: [constraints]
-Group [X]: [group name] - Subtask 1 of N in this group
-" --skip-git-repo-check -s danger-full-access
-```
-
-**For Related Subtasks in Same Group** (Resume Session):
-```bash
-# Stage changes from previous subtask (if valid git repository)
-git add -A
-
-# Resume session ONLY for subtasks in same group
-codex exec "
-CONTINUE IN SAME GROUP:
-Group [X]: [group name] - Subtask N of M
-
-PURPOSE: [continuation goal within group]
-TASK: [subtask N description]
-CONTEXT: Previous work in this group completed, now focus on @{new_relevant_files}
-EXPECTED: [specific deliverables]
-RULES: Build on previous subtask in group, maintain consistency
-" resume --last --skip-git-repo-check -s danger-full-access
-```
-
-**For First Subtask in Different Group** (New Session):
-```bash
-# Stage changes from previous group
-git add -A
-
-# Start NEW session for different group (no resume)
-codex -C [dir] --full-auto exec "
-PURPOSE: [new group goal]
-TASK: [subtask description - first in new group]
-CONTEXT: @{different_files} @CLAUDE.md
-EXPECTED: [specific deliverables]
-RULES: [constraints]
-Group [Y]: [new group name] - Subtask 1 of N in this group
-" --skip-git-repo-check -s danger-full-access
-```
-
-**Resume Decision Logic**:
-```
-if (subtask.group == previous_subtask.group):
-    use `codex exec "..." resume --last`  # Continue session
-else:
-    use `codex -C [dir] exec "..."`       # New session
-```
-
-### Phase 3: Verification (if --verify-git enabled)
-
-After each subtask completion:
-```bash
-# Check git status
-git status --short
-
-# Verify expected changes
-git diff --stat
-
-# Optional: Check for untracked files that should be committed
-git ls-files --others --exclude-standard
-```
-
-**Verification Checks**:
-- Files modified match subtask scope
-- No unexpected changes in unrelated files
-- No merge conflicts or errors
-- Code compiles/runs (if applicable)
-
-### Phase 4: TodoWrite Tracking with Groups
-
-**Initial Setup with Task Flow**:
-```javascript
-TodoWrite({
-  todos: [
-    // Display task flow diagram first
-    { content: "Task Flow Analysis Complete - See diagram above", status: "completed", activeForm: "Analyzing task flow" },
-
-    // Group A subtasks (will use resume within group)
-    { content: "[Group A] Subtask 1: [description]", status: "in_progress", activeForm: "Executing Group A subtask 1" },
-    { content: "[Group A] Subtask 2: [description] [resume]", status: "pending", activeForm: "Executing Group A subtask 2" },
-
-    // Group B subtasks (new session, then resume within group)
-    { content: "[Group B] Subtask 1: [description] [new session]", status: "pending", activeForm: "Executing Group B subtask 1" },
-    { content: "[Group B] Subtask 2: [description] [resume]", status: "pending", activeForm: "Executing Group B subtask 2" },
-
-    // Group C subtasks (new session)
-    { content: "[Group C] Subtask 1: [description] [new session]", status: "pending", activeForm: "Executing Group C subtask 1" },
-
-    { content: "Final verification and summary", status: "pending", activeForm: "Verifying and summarizing" }
-  ]
-})
-```
-
-**After Each Subtask**:
-```javascript
-TodoWrite({
-  todos: [
-    { content: "Task Flow Analysis Complete - See diagram above", status: "completed", activeForm: "Analyzing task flow" },
-    { content: "[Group A] Subtask 1: [description]", status: "completed", activeForm: "Executing Group A subtask 1" },
-    { content: "[Group A] Subtask 2: [description] [resume]", status: "in_progress", activeForm: "Executing Group A subtask 2" },
-    // ... update status
-  ]
-})
-```
-
-## Codex Resume Mechanism
-
-**Why Group-Based Resume?**
-- **Within Group**: Maintains conversation context for related subtasks
-  - Codex remembers previous decisions and patterns
-  - Reduces context repetition
-  - Ensures consistency in implementation style
-- **Between Groups**: Fresh session for independent tasks
-  - Avoids context pollution from unrelated work
-  - Prevents confusion when switching domains
-  - Maintains focused attention on current group
-
-**How It Works**:
-1. **First subtask in Group A**: Creates new Codex session
-2. **Subsequent subtasks in Group A**: Use `codex resume --last` to continue session
-3. **First subtask in Group B**: Creates NEW Codex session (no resume)
-4. **Subsequent subtasks in Group B**: Use `codex resume --last` within Group B
-5. Each group builds on its own context, isolated from other groups
-
-**When to Resume vs New Session**:
-```
-RESUME (same group):
-  - Subtasks share files/modules
-  - Logical continuation of previous work
-  - Same architectural domain
-
-NEW SESSION (different group):
-  - Independent task area
-  - Different files/modules
-  - Switching architectural domains
-  - Testing after implementation
-```
-
-**Image Support**:
-```bash
-# First subtask with design reference
-codex -C [dir] -i design.png --full-auto exec "..." --skip-git-repo-check -s danger-full-access
-
-# Resume for next subtask (image context preserved)
-codex exec "CONTINUE TO NEXT SUBTASK: ..." resume --last --skip-git-repo-check -s danger-full-access
-```
-
-## Error Handling
-
-**Subtask Failure**:
-1. Mark subtask as blocked in TodoWrite
-2. Report error details to user
-3. Pause execution for manual intervention
-4. Use AskUserQuestion for recovery decision:
-
-```typescript
-AskUserQuestion({
-  questions: [{
-    question: "Codex execution failed for the subtask. How should the workflow proceed?",
-    header: "Recovery",
-    options: [
-      { label: "Retry Subtask", description: "Attempt to execute the same subtask again." },
-      { label: "Skip Subtask", description: "Continue to the next subtask in the plan." },
-      { label: "Abort Workflow", description: "Stop the entire execution." }
-    ],
-    multiSelect: false
-  }]
-})
-```
-
-**Git Verification Failure** (if --verify-git):
-1. Show unexpected changes
-2. Pause execution
-3. Request user decision:
-   - Continue anyway
-   - Rollback and retry
-   - Manual fix
-
-**Codex Session Lost**:
-1. Detect if `codex exec "..." resume --last` fails
-2. Attempt retry with fresh session
-3. Report to user if manual intervention needed
-
-## Output Format
-
-**During Execution**:
-```
-Task Flow Diagram:
-[Group A: Auth Core]
-  A1: Create user model ──┐
-  A2: Add validation     ─┤─► [resume] ─► A3: Database schema
-                          │
-[Group B: API Layer]      │
-  B1: Auth endpoints ─────┘─► [new session]
-  B2: Middleware ────────────► [resume] ─► B3: Error handling
-
-[Group C: Testing]
-  C1: Unit tests ─────────────► [new session]
-  C2: Integration tests ──────► [resume]
-
-Task Decomposition:
-  [Group A] 1. Create user model
-  [Group A] 2. Add validation logic [resume]
-  [Group A] 3. Implement database schema [resume]
-  [Group B] 4. Create auth endpoints [new session]
-  [Group B] 5. Add middleware [resume]
-  [Group B] 6. Error handling [resume]
-  [Group C] 7. Unit tests [new session]
-  [Group C] 8. Integration tests [resume]
-
-[Group A] Executing Subtask 1/8: Create user model
-  Starting new Codex session for Group A...
-  [Codex output]
-  Subtask 1 completed
-
-Git Verification:
-  M  src/models/user.ts
-  Changes verified
-
-[Group A] Executing Subtask 2/8: Add validation logic
-  Resuming Codex session (same group)...
-  [Codex output]
-  Subtask 2 completed
-
-[Group B] Executing Subtask 4/8: Create auth endpoints
-  Starting NEW Codex session for Group B...
-  [Codex output]
-  Subtask 4 completed
-...
-
-All Subtasks Completed
-Summary: [file references, changes, next steps]
-```
-
-**Final Summary**:
-```markdown
-# Task Execution Summary: [Task Description]
-
-## Subtasks Completed
-1. [Subtask 1]: [files modified]
-2. [Subtask 2]: [files modified]
-...
-
-## Files Modified
-- src/file1.ts:10-50 - [changes]
-- src/file2.ts - [changes]
-
-## Git Status
-- N files modified
-- M files added
-- No conflicts
-
-## Next Steps
-- [Suggested follow-up actions]
-```
-
-## Examples
-
-**Example 1: Simple Task with Groups**
-```bash
-/cli:codex-execute "implement user authentication system"
-
-# Task Flow Diagram:
-# [Group A: Data Layer]
-#   A1: Create user model ──► [resume] ──► A2: Database schema
-#
-# [Group B: Auth Logic]
-#   B1: JWT token generation ──► [new session]
-#   B2: Authentication middleware ──► [resume]
-#
-# [Group C: API Endpoints]
-#   C1: Login/logout endpoints ──► [new session]
-#
-# [Group D: Testing]
-#   D1: Unit tests ──► [new session]
-#   D2: Integration tests ──► [resume]
-
-# Execution:
-# Group A: A1 (new) → A2 (resume)
-# Group B: B1 (new) → B2 (resume)
-# Group C: C1 (new)
-# Group D: D1 (new) → D2 (resume)
-```
-
-**Example 2: With Git Verification**
-```bash
-/cli:codex-execute --verify-git "refactor API layer to use dependency injection"
-
-# After each subtask, verifies:
-# - Only expected files modified
-# - No breaking changes in unrelated code
-# - Tests still pass
-```
-
-**Example 3: With Task ID**
-```bash
-/cli:codex-execute IMPL-001
-
-# Loads task from .task/IMPL-001.json
-# Decomposes based on task requirements
-```
-
-## Best Practices
-
-1. **Task Flow First**: Always create visual flow diagram before execution
-2. **Group Related Work**: Cluster subtasks by domain/files for efficient resume
-3. **Subtask Granularity**: Keep subtasks small and focused (5-15 min each)
-4. **Clear Boundaries**: Each subtask should have well-defined input/output
-5. **Git Hygiene**: Use `--verify-git` for critical refactoring
-6. **Pre-Execution Staging**: Stage changes before each subtask to clearly see codex modifications
-7. **Smart Resume**: Use `resume --last` ONLY within same group
-8. **Fresh Sessions**: Start new session when switching to different group/domain
-9. **Recovery Points**: TodoWrite with group labels provides clear progress tracking
-10. **Image References**: Attach design files for UI tasks (first subtask in group)
-
-## Input Processing
-
-**Automatic Detection**:
-- Input matches task ID pattern → Load from `.task/[ID].json`
-- Otherwise → Use as task description
-
-**Task JSON Structure** (when loading from file):
-```json
-{
-  "task_id": "IMPL-001",
-  "title": "Implement user authentication",
-  "description": "Create JWT-based auth system",
-  "acceptance_criteria": [...],
-  "scope": {...},
-  "brainstorming_refs": [...]
-}
-```
-
-## Output Routing
-
-**Execution Log Destination**:
-- **IF** active workflow session exists:
-  - Execution log: `.workflow/WFS-[id]/.chat/codex-execute-[timestamp].md`
-  - Task summaries: `.workflow/WFS-[id]/.summaries/[TASK-ID]-summary.md` (if task ID)
-  - Task updates: `.workflow/WFS-[id]/.task/[TASK-ID].json` status updates
-  - TodoWrite tracking: Embedded in execution log
-- **ELSE** (no active session):
-  - **Recommended**: Create workflow session first (`/workflow:session:start`)
-  - **Alternative**: Save to `.workflow/.scratchpad/codex-execute-[description]-[timestamp].md`
-
-**Output Files** (during execution):
-```
-.workflow/WFS-[session-id]/
-├── .chat/
-│   └── codex-execute-20250105-143022.md    # Full execution log with task flow
-├── .summaries/
-│   ├── IMPL-001.1-summary.md               # Subtask summaries
-│   ├── IMPL-001.2-summary.md
-│   └── IMPL-001-summary.md                 # Final task summary
-└── .task/
-    ├── IMPL-001.json                       # Updated task status
-    └── [subtask JSONs if decomposed]
-```
-
-**Examples**:
-- During session `WFS-auth-system`, executing multi-stage auth implementation:
-  - Log: `.workflow/WFS-auth-system/.chat/codex-execute-20250105-143022.md`
-  - Summaries: `.workflow/WFS-auth-system/.summaries/IMPL-001.{1,2,3}-summary.md`
-  - Task status: `.workflow/WFS-auth-system/.task/IMPL-001.json` (status: completed)
-- No session, ad-hoc multi-stage task:
-  - Log: `.workflow/.scratchpad/codex-execute-auth-refactor-20250105-143045.md`
-
-**Save Results**:
-- Execution log with task flow diagram and TodoWrite tracking
-- Individual summaries for each completed subtask
-- Final consolidated summary when all subtasks complete
-- Modified code files throughout project
-
-## Notes
-
-**vs. `/cli:execute`**:
-- `/cli:execute`: Single-shot execution with Gemini/Qwen/Codex
-- `/cli:codex-execute`: Multi-stage Codex execution with automatic task decomposition and resume mechanism
-
-**Input Flexibility**: Accepts both freeform descriptions and task IDs (auto-detects and loads JSON)
-
-**Context Window**: `codex exec "..." resume --last` maintains conversation history, ensuring consistency across subtasks without redundant context injection.
-
-**Output Details**:
-- Session management: see intelligent-tools-strategy.md
-- **⚠️ Code Modification**: This command performs multi-stage code modifications - execution log tracks all changes

.claude/commands/cli/discuss-plan.md

@@ -1,320 +0,0 @@
----
-name: discuss-plan
-description: Orchestrates an iterative, multi-model discussion for planning and analysis without implementation.
-argument-hint: "[--topic '...'] [--task-id '...'] [--rounds N]"
-allowed-tools: SlashCommand(*), Bash(*), TodoWrite(*), Read(*), Glob(*)
----
-
-# CLI Discuss-Plan Command (/cli:discuss-plan)
-
-## Purpose
-
-Orchestrates a multi-model collaborative discussion for in-depth planning and problem analysis. This command facilitates an iterative dialogue between Gemini, Codex, and Claude (the orchestrating AI) to explore a topic from multiple perspectives, refine ideas, and build a robust plan.
-
-**This command is for discussion and planning ONLY. It does NOT modify any code.**
-
-## Core Workflow: The Discussion Loop
-
-The command operates in iterative rounds, allowing the plan to evolve with each cycle. The user can choose to continue for more rounds or conclude when consensus is reached.
-
-```
-Topic Input → [Round 1: Gemini → Codex → Claude] → [User Review] →
-[Round 2: Gemini → Codex → Claude] → ... → Final Plan
-```
-
-### Model Roles & Priority
-
-**Priority Order**: Gemini > Codex > Claude
-
-1.  **Gemini (The Analyst)** - Priority 1
-    - Kicks off each round with deep analysis
-    - Provides foundational ideas and draft plans
-    - Analyzes current context or previous synthesis
-
-2.  **Codex (The Architect/Critic)** - Priority 2
-    - Reviews Gemini's output critically
-    - Uses deep reasoning for technical trade-offs
-    - Proposes alternative strategies
-    - **Participates purely in conversational/reasoning capacity**
-    - Uses resume mechanism to maintain discussion context
-
-3.  **Claude (The Synthesizer/Moderator)** - Priority 3
-    - Synthesizes discussion from Gemini and Codex
-    - Highlights agreements and contentions
-    - Structures refined plan
-    - Poses key questions for next round
-
-## Parameters
-
--   `<input>` (Required): Topic description or task ID (e.g., "Design a new caching layer" or `PLAN-002`)
--   `--rounds <N>` (Optional): Maximum number of discussion rounds (default: prompts after each round)
--   `--task-id <id>` (Optional): Associates discussion with workflow task ID
--   `--topic <description>` (Optional): High-level topic for discussion
-
-## Execution Flow
-
-### Phase 1: Initial Setup
-
-1.  **Input Processing**: Parse topic or task ID
-2.  **Context Gathering**: Identify relevant files based on topic
-
-### Phase 2: Discussion Round
-
-Each round consists of three sequential steps, tracked via `TodoWrite`.
-
-**Step 1: Gemini's Analysis (Priority 1)**
-
-Gemini analyzes the topic and proposes preliminary plan.
-
-```bash
-# Round 1: CONTEXT_INPUT is the initial topic
-# Subsequent rounds: CONTEXT_INPUT is the synthesis from previous round
-gemini -p "
-PURPOSE: Analyze and propose a plan for '[topic]'
-TASK: Provide initial analysis, identify key modules, and draft implementation plan
-MODE: analysis
-CONTEXT: @CLAUDE.md [auto-detected files]
-INPUT: [CONTEXT_INPUT]
-EXPECTED: Structured analysis and draft plan for discussion
-RULES: Focus on technical depth and practical considerations
-"
-```
-
-**Step 2: Codex's Critique (Priority 2)**
-
-Codex reviews Gemini's output using conversational reasoning. Uses `resume --last` to maintain context across rounds.
-
-```bash
-# First round (new session)
-codex --full-auto exec "
-PURPOSE: Critically review technical plan
-TASK: Review the provided plan, identify weaknesses, suggest alternatives, reason about trade-offs
-MODE: analysis
-CONTEXT: @CLAUDE.md [relevant files]
-INPUT_PLAN: [Output from Gemini's analysis]
-EXPECTED: Critical review with alternative ideas and risk analysis
-RULES: Focus on architectural soundness and implementation feasibility
-" --skip-git-repo-check
-
-# Subsequent rounds (resume discussion)
-codex --full-auto exec "
-PURPOSE: Re-evaluate plan based on latest synthesis
-TASK: Review updated plan and discussion points, provide further critique or refined ideas
-MODE: analysis
-CONTEXT: Previous discussion context (maintained via resume)
-INPUT_PLAN: [Output from Gemini's analysis for current round]
-EXPECTED: Updated critique building on previous discussion
-RULES: Build on previous insights, avoid repeating points
-" resume --last --skip-git-repo-check
-```
-
-**Step 3: Claude's Synthesis (Priority 3)**
-
-Claude (orchestrating AI) synthesizes both outputs:
-
--   Summarizes Gemini's proposal and Codex's critique
--   Highlights agreements and disagreements
--   Structures consolidated plan
--   Presents open questions for next round
--   This synthesis becomes input for next round
-
-### Phase 3: User Review and Iteration
-
-1.  **Present Synthesis**: Show synthesized plan and key discussion points
-2.  **Continue or Conclude**: Use AskUserQuestion to prompt user:
-
-```typescript
-AskUserQuestion({
-  questions: [{
-    question: "Round of discussion complete. What is the next step?",
-    header: "Next Round",
-    options: [
-      { label: "Start another round", description: "Continue the discussion to refine the plan further." },
-      { label: "Conclude and finalize", description: "End the discussion and save the final plan." }
-    ],
-    multiSelect: false
-  }]
-})
-```
-
-3.  **Loop or Finalize**:
-    -   Continue → New round with Gemini analyzing latest synthesis
-    -   Conclude → Save final synthesized document
-
-## TodoWrite Tracking
-
-Progress tracked for each round and model.
-
-```javascript
-// Example for 2-round discussion
-TodoWrite({
-  todos: [
-    // Round 1
-    { content: "[Round 1] Gemini: Analyzing topic", status: "completed", activeForm: "Analyzing with Gemini" },
-    { content: "[Round 1] Codex: Critiquing plan", status: "completed", activeForm: "Critiquing with Codex" },
-    { content: "[Round 1] Claude: Synthesizing discussion", status: "completed", activeForm: "Synthesizing discussion" },
-    { content: "[User Action] Review Round 1 and decide next step", status: "in_progress", activeForm: "Awaiting user decision" },
-
-    // Round 2
-    { content: "[Round 2] Gemini: Analyzing refined plan", status: "pending", activeForm: "Analyzing refined plan" },
-    { content: "[Round 2] Codex: Re-evaluating plan [resume]", status: "pending", activeForm: "Re-evaluating with Codex" },
-    { content: "[Round 2] Claude: Finalizing plan", status: "pending", activeForm: "Finalizing plan" },
-    { content: "Discussion complete - Final plan generated", status: "pending", activeForm: "Generating final document" }
-  ]
-})
-```
-
-## Output Routing
-
--   **Primary Log**: Entire multi-round discussion logged to single file:
-    -   `.workflow/WFS-[id]/.chat/discuss-plan-[topic]-[timestamp].md`
--   **Final Plan**: Clean final version saved upon conclusion:
-    -   `.workflow/WFS-[id]/.summaries/plan-[topic].md`
--   **Scratchpad**: If no session active:
-    -   `.workflow/.scratchpad/discuss-plan-[topic]-[timestamp].md`
-
-## Discussion Structure
-
-Each round's output is structured as:
-
-```markdown
-## Round N: [Topic]
-
-### Gemini's Analysis (Priority 1)
-[Gemini's full analysis and proposal]
-
-### Codex's Critique (Priority 2)
-[Codex's critical review and alternatives]
-
-### Claude's Synthesis (Priority 3)
-**Points of Agreement:**
-- [Agreement 1]
-- [Agreement 2]
-
-**Points of Contention:**
-- [Issue 1]: Gemini suggests X, Codex suggests Y
-- [Issue 2]: Trade-off between A and B
-
-**Consolidated Plan:**
-[Structured plan incorporating both perspectives]
-
-**Open Questions for Next Round:**
-1. [Question 1]
-2. [Question 2]
-```
-
-## Examples
-
-### Example 1: Multi-Round Architecture Discussion
-
-**Command**: `/cli:discuss-plan --topic "Design a real-time notification system"`
-
-**Round 1**:
-1.  **Gemini**: Proposes WebSocket-based architecture with RabbitMQ message queue
-2.  **Codex**: Critiques as overly complex for MVP. Suggests Server-Sent Events (SSE) for simplicity (one-way notifications). Questions RabbitMQ necessity, proposes simpler Redis Pub/Sub
-3.  **Claude**: Synthesizes views:
-    -   **Plan A (Gemini)**: WebSockets + RabbitMQ (highly scalable, complex)
-    -   **Plan B (Codex)**: SSE + Redis (simpler, less overhead)
-    -   **Open Question**: Is bi-directional communication critical, or is simplicity priority?
-4.  **User Action**: Opts for another round to explore trade-offs
-
-**Round 2**:
-1.  **Gemini**: Analyzes synthesized document. Notes that if features like "user is typing" indicators are roadmapped, WebSockets better long-term. Drafts plan starting with SSE/Redis but designing for easy migration
-2.  **Codex**: Reviews migration plan. Reasons that migration itself could be complex. If feature set likely to expand, starting with WebSockets using managed service might be best cost/benefit
-3.  **Claude**: Synthesizes new discussion:
-    -   **Consensus**: Simple SSE/Redis too short-sighted
-    -   **Refined Options**:
-        1.  Phased approach (SSE → WebSocket) with clear migration plan
-        2.  Direct WebSocket with managed service (Pusher, Ably) to reduce ops overhead
-    -   **Recommendation**: Option 2 most robust and future-proof
-4.  **User Action**: Agrees with recommendation, concludes discussion
-
-**Final Output**: Planning document saved with:
-- Chosen architecture (Managed WebSocket service)
-- Multi-round reasoning
-- High-level implementation steps
-
-### Example 2: Feature Design Discussion
-
-**Command**: `/cli:discuss-plan --topic "Design user permission system" --rounds 2`
-
-**Round 1**:
-1.  **Gemini**: Proposes RBAC (Role-Based Access Control) with predefined roles
-2.  **Codex**: Suggests ABAC (Attribute-Based Access Control) for more flexibility
-3.  **Claude**: Synthesizes trade-offs between simplicity (RBAC) vs flexibility (ABAC)
-
-**Round 2**:
-1.  **Gemini**: Analyzes hybrid approach - RBAC for core permissions, attributes for fine-grained control
-2.  **Codex**: Reviews hybrid model, identifies implementation challenges
-3.  **Claude**: Final plan with phased rollout strategy
-
-**Automatic Conclusion**: Command concludes after 2 rounds as specified
-
-### Example 3: Problem-Solving Discussion
-
-**Command**: `/cli:discuss-plan --topic "Debug memory leak in data pipeline" --task-id ISSUE-042`
-
-**Round 1**:
-1.  **Gemini**: Identifies potential leak sources (unclosed handles, growing cache, event listeners)
-2.  **Codex**: Adds profiling tool recommendations, suggests memory monitoring
-3.  **Claude**: Structures debugging plan with phased approach
-
-**User Decision**: Single round sufficient, concludes with debugging strategy
-
-## Consensus Mechanisms
-
-**When to Continue:**
-- Significant disagreement between models
-- Open questions requiring deeper analysis
-- Trade-offs need more exploration
-- User wants additional perspectives
-
-**When to Conclude:**
-- Models converge on solution
-- All key questions addressed
-- User satisfied with plan depth
-- Maximum rounds reached (if specified)
-
-## Comparison with Other Commands
-
-| Command | Models | Rounds | Discussion | Implementation | Use Case |
-|---------|--------|--------|------------|----------------|----------|
-| `/cli:mode:plan` | Gemini | 1 | NO | NO | Single-model planning |
-| `/cli:analyze` | Gemini/Qwen | 1 | NO | NO | Code analysis |
-| `/cli:execute` | Any | 1 | NO | YES | Direct implementation |
-| `/cli:codex-execute` | Codex | 1 | NO | YES | Multi-stage implementation |
-| `/cli:discuss-plan` | **Gemini+Codex+Claude** | **Multiple** | **YES** | **NO** | **Multi-perspective planning** |
-
-## Best Practices
-
-1.  **Use for Complex Decisions**: Ideal for architectural decisions, design trade-offs, problem-solving
-2.  **Start with Broad Topic**: Let first round establish scope, subsequent rounds refine
-3.  **Review Each Synthesis**: Claude's synthesis is key decision point - review carefully
-4.  **Know When to Stop**: Don't over-iterate - 2-3 rounds usually sufficient
-5.  **Task Association**: Use `--task-id` for traceability in workflow
-6.  **Save Intermediate Results**: Each round's synthesis saved automatically
-7.  **Let Models Disagree**: Divergent views often reveal important trade-offs
-8.  **Focus Questions**: Use Claude's open questions to guide next round
-
-## Breaking Discussion Loops
-
-**Detecting Loops:**
-- Models repeating same arguments
-- No new insights emerging
-- Trade-offs well understood
-
-**Breaking Strategies:**
-1.  **User Decision**: Make executive decision when enough info gathered
-2.  **Timeboxing**: Set max rounds upfront with `--rounds`
-3.  **Criteria-Based**: Define decision criteria before starting
-4.  **Hybrid Approach**: Accept multiple valid solutions in final plan
-
-## Notes
-
--   **Pure Discussion**: This command NEVER modifies code - only produces planning documents
--   **Codex Role**: Codex participates as reasoning/critique tool, not executor
--   **Resume Context**: Codex maintains discussion context via `resume --last`
--   **Priority System**: Ensures Gemini leads analysis, Codex provides critique, Claude synthesizes
--   **Output Quality**: Multi-perspective discussion produces more robust plans than single-model analysis
--   Command patterns and session management: see intelligent-tools-strategy.md (loaded in memory)
--   For implementation after discussion, use `/cli:execute` or `/cli:codex-execute` separately

.claude/commands/cli/execute.md

@@ -1,221 +0,0 @@
----
-name: execute
-description: Auto-execution of implementation tasks with YOLO permissions and intelligent context inference
-argument-hint: "[--agent] [--tool codex|gemini|qwen] [--enhance] description or task-id"
-allowed-tools: SlashCommand(*), Bash(*), Task(*)
----
-
-# CLI Execute Command (/cli:execute)
-
-## Purpose
-
-Execute implementation tasks with **YOLO permissions** (auto-approves all confirmations). **MODIFIES CODE**.
-
-**Intent**: Autonomous code implementation, modification, and generation
-**Supported Tools**: codex, gemini (default), qwen
-**Key Feature**: Automatic context inference and file pattern detection
-
-## Core Behavior
-
-1. **Code Modification**: This command MODIFIES, CREATES, and DELETES code files
-2. **Auto-Approval**: YOLO mode bypasses confirmation prompts for all operations
-3. **Implementation Focus**: Executes actual code changes, not just recommendations
-4. **Requires Explicit Intent**: Use only when implementation is intended
-
-## Core Concepts
-
-### YOLO Permissions
-Auto-approves: file pattern inference, execution, **file modifications**, summary generation
-
-**WARNING**: This command will make actual code changes without manual confirmation
-
-### Execution Modes
-
-**1. Description Mode** (supports `--enhance`):
-- Input: Natural language description
-- Process: [Optional: Enhance] → Keyword analysis → Pattern inference → Execute
-
-**2. Task ID Mode** (no `--enhance`):
-- Input: Workflow task identifier (e.g., `IMPL-001`)
-- Process: Task JSON parsing → Scope analysis → Execute
-
-**3. Agent Mode** (`--agent` flag):
-- Input: Description or task-id
-- Process: 5-Phase Workflow → Context Discovery → Optimal Tool Selection → Execute
-
-### Context Inference
-
-Auto-selects files based on keywords and technology (each @ references one pattern):
-- "auth" → `@**/*auth* @**/*user*`
-- "React" → `@src/**/*.jsx @src/**/*.tsx`
-- "api" → `@**/api/**/* @**/routes/**/*`
-- Always includes: `@CLAUDE.md @**/*CLAUDE.md`
-
-For precise file targeting, use `rg` or MCP tools to discover files first.
-
-### Codex Session Continuity
-
-**Resume Pattern** for related tasks:
-```bash
-# First task - establish session
-codex -C [dir] --full-auto exec "[task]" --skip-git-repo-check -s danger-full-access
-
-# Related task - continue session
-codex --full-auto exec "[related-task]" resume --last --skip-git-repo-check -s danger-full-access
-```
-
-Use `resume --last` when current task extends/relates to previous execution. See intelligent-tools-strategy.md for auto-resume rules.
-
-## Parameters
-
-- `--agent` - Use cli-execution-agent for automated context discovery (5-phase intelligent mode)
-- `--tool <codex|gemini|qwen>` - Select CLI tool (default: gemini, ignored in agent mode unless specified)
-- `--enhance` - Enhance input with `/enhance-prompt` first (Description Mode only)
-- `<description|task-id>` - Natural language description or task identifier
-- `--debug` - Verbose logging
-- `--save-session` - Save execution to workflow session
-
-## Workflow Integration
-
-**Session Management**: Auto-detects `.workflow/.active-*` marker
-- Active session: Save to `.workflow/WFS-[id]/.chat/execute-[timestamp].md`
-- No session: Create new session or save to scratchpad
-
-**Task Integration**: Load from `.task/[TASK-ID].json`, update status, generate summary
-
-## Output Routing
-
-**Execution Log Destination**:
-- **IF** active workflow session exists:
-  - Save to `.workflow/WFS-[id]/.chat/execute-[timestamp].md`
-  - Update task status in `.task/[TASK-ID].json` (if task ID provided)
-  - Generate summary in `.workflow/WFS-[id]/.summaries/[TASK-ID]-summary.md`
-- **ELSE** (no active session):
-  - **Option 1**: Create new workflow session for task
-  - **Option 2**: Save to `.workflow/.scratchpad/execute-[description]-[timestamp].md`
-
-**Output Files** (when active session exists):
-- Execution log: `.workflow/WFS-[id]/.chat/execute-[timestamp].md`
-- Task summary: `.workflow/WFS-[id]/.summaries/[TASK-ID]-summary.md` (if task ID)
-- Modified code: Project files per implementation
-
-**Examples**:
-- During session `WFS-auth-system`, executing `IMPL-001`:
-  - Log: `.workflow/WFS-auth-system/.chat/execute-20250105-143022.md`
-  - Summary: `.workflow/WFS-auth-system/.summaries/IMPL-001-summary.md`
-- No session, ad-hoc implementation:
-  - Log: `.workflow/.scratchpad/execute-jwt-auth-20250105-143045.md`
-
-## Execution Modes
-
-### Standard Mode (Default)
-```bash
-# Gemini/Qwen: MODE=write with --approval-mode yolo
-cd . && gemini --approval-mode yolo "
-PURPOSE: [implementation goal]
-TASK: [specific implementation]
-MODE: write
-CONTEXT: @CLAUDE.md [auto-detected files]
-EXPECTED: Working implementation with code changes
-RULES: [constraints] | Auto-approve all changes
-"
-
-# Codex: MODE=auto with danger-full-access
-codex -C . --full-auto exec "
-PURPOSE: [implementation goal]
-TASK: [specific implementation]
-MODE: auto
-CONTEXT: [auto-detected files]
-EXPECTED: Complete implementation with tests
-" --skip-git-repo-check -s danger-full-access
-```
-
-### Agent Mode (`--agent` flag)
-
-Delegate implementation to `cli-execution-agent` for intelligent execution with automated context discovery.
-
-**Agent invocation**:
-```javascript
-Task(
-  subagent_type="cli-execution-agent",
-  description="Implement with automated context discovery and optimal tool selection",
-  prompt=`
-    Task: ${description_or_task_id}
-    Mode: execute
-    Tool Preference: ${tool_flag || 'auto-select'}
-    ${enhance_flag ? 'Enhance: true' : ''}
-
-    Agent will autonomously:
-    - Discover implementation files and dependencies
-    - Assess complexity and select optimal tool
-    - Execute with YOLO permissions (auto-approve)
-    - Generate task summary if task-id provided
-  `
-)
-```
-
-The agent handles all phases internally, including complexity-based tool selection.
-
-## Examples
-
-**Basic Implementation (Standard Mode)** (modifies code):
-```bash
-/cli:execute "implement JWT authentication with middleware"
-# Executes: Creates auth middleware, updates routes, modifies config
-# Result: NEW/MODIFIED code files with JWT implementation
-```
-
-**Intelligent Implementation (Agent Mode)** (modifies code):
-```bash
-/cli:execute --agent "implement OAuth2 authentication with token refresh"
-# Phase 1: Classifies intent=execute, complexity=complex, keywords=['oauth2', 'auth', 'token', 'refresh']
-# Phase 2: MCP discovers auth patterns, existing middleware, JWT dependencies
-# Phase 3: Enhances prompt with discovered patterns and best practices
-# Phase 4: Selects Codex (complex task), executes with comprehensive context
-# Phase 5: Saves execution log + generates implementation summary
-# Result: Complete OAuth2 implementation + detailed execution log
-```
-
-**Enhanced Implementation** (modifies code):
-```bash
-/cli:execute --enhance "implement JWT authentication"
-# Step 1: Enhance to expand requirements
-# Step 2: Execute implementation with auto-approval
-# Result: Complete auth system with MODIFIED code files
-```
-
-**Task Execution** (modifies code):
-```bash
-/cli:execute IMPL-001
-# Reads: .task/IMPL-001.json for requirements
-# Executes: Implementation based on task spec
-# Result: Code changes per task definition
-```
-
-**Codex Implementation** (modifies code):
-```bash
-/cli:execute --tool codex "optimize database queries"
-# Executes: Codex with full file access
-# Result: MODIFIED query code, new indexes, updated tests
-```
-
-**Qwen Code Generation** (modifies code):
-```bash
-/cli:execute --tool qwen --enhance "refactor auth module"
-# Step 1: Enhanced refactoring plan
-# Step 2: Execute with MODE=write
-# Result: REFACTORED auth code with structural changes
-```
-
-## Comparison with Analysis Commands
-
-| Command | Intent | Code Changes | Auto-Approve |
-|---------|--------|--------------|--------------|
-| `/cli:analyze` | Understand code | NO | N/A |
-| `/cli:chat` | Ask questions | NO | N/A |
-| `/cli:execute` | **Implement** | **YES** | **YES** |
-
-## Notes
-
-- Command templates, YOLO mode details, and session management: see intelligent-tools-strategy.md (loaded in memory)
-- **Code Modification**: This command modifies code - execution logs document changes made

.claude/commands/cli/mode/bug-diagnosis.md

@@ -1,130 +0,0 @@
----
-name: bug-diagnosis
-description: Bug diagnosis and fix suggestions using CLI tools with specialized template
-argument-hint: "[--agent] [--tool codex|gemini|qwen] [--enhance] [--cd path] bug description"
-allowed-tools: SlashCommand(*), Bash(*), Task(*)
----
-
-# CLI Mode: Bug Diagnosis (/cli:mode:bug-diagnosis)
-
-## Purpose
-
-Systematic bug diagnosis with root cause analysis template (`~/.claude/workflows/cli-templates/prompts/analysis/01-diagnose-bug-root-cause.txt`).
-
-**Tool Selection**:
-- **gemini** (default) - Best for bug diagnosis
-- **qwen** - Fallback when Gemini unavailable
-- **codex** - Alternative for complex bug analysis
-
-## Parameters
-
-- `--tool <gemini|qwen|codex>` - Tool selection (default: gemini)
-- `--agent` - Use cli-execution-agent for automated context discovery
-- `--enhance` - Enhance bug description with `/enhance-prompt`
-- `--cd "path"` - Target directory for focused diagnosis
-- `<bug-description>` (Required) - Bug description or error details
-
-## Tool Usage
-
-**Gemini** (Primary):
-```bash
-# Uses gemini by default, or specify explicitly
---tool gemini
-```
-
-**Qwen** (Fallback):
-```bash
---tool qwen
-```
-
-**Codex** (Alternative):
-```bash
---tool codex
-```
-
-## Execution Flow
-
-### Standard Mode
-1. Parse tool selection (default: gemini)
-2. Optional: enhance with `/enhance-prompt`
-3. Detect directory from `--cd` or auto-infer
-4. Build command with template
-5. Execute diagnosis (read-only)
-6. Save to `.workflow/WFS-[id]/.chat/`
-
-### Agent Mode (`--agent`)
-
-Delegates to agent for intelligent diagnosis:
-
-```javascript
-Task(
-  subagent_type="cli-execution-agent",
-  description="Bug root cause diagnosis",
-  prompt=`
-    Task: ${bug_description}
-    Mode: bug-diagnosis
-    Tool: ${tool_flag || 'auto-select'}  // gemini|qwen|codex
-    Directory: ${cd_path || 'auto-detect'}
-    Template: ~/.claude/workflows/cli-templates/prompts/analysis/01-diagnose-bug-root-cause.txt
-
-    Agent responsibilities:
-    1. Context Discovery:
-       - Locate error traces and logs
-       - Find related code sections
-       - Identify data flow paths
-
-    2. CLI Command Generation:
-       - Build Gemini/Qwen/Codex command
-       - Include diagnostic context
-       - Apply ~/.claude/workflows/cli-templates/prompts/analysis/01-diagnose-bug-root-cause.txt template
-
-    3. Execution & Output:
-       - Execute root cause analysis
-       - Generate fix suggestions
-       - Save to .workflow/.chat/
-  `
-)
-```
-
-## Core Rules
-
-- **Read-only**: Diagnoses bugs, does NOT modify code
-- **Template**: Uses `~/.claude/workflows/cli-templates/prompts/analysis/01-diagnose-bug-root-cause.txt` for root cause analysis
-- **Output**: Saves to `.workflow/WFS-[id]/.chat/`
-
-## CLI Command Templates
-
-**Gemini/Qwen** (default, diagnosis only):
-```bash
-cd [dir] && gemini -p "
-PURPOSE: [goal]
-TASK: Root cause analysis
-MODE: analysis
-CONTEXT: @**/*
-EXPECTED: Diagnosis, fix plan
-RULES: $(cat ~/.claude/workflows/cli-templates/prompts/analysis/01-diagnose-bug-root-cause.txt)
-"
-# Qwen: Replace 'gemini' with 'qwen'
-```
-
-**Codex** (diagnosis + potential fixes):
-```bash
-codex -C [dir] --full-auto exec "
-PURPOSE: [goal]
-TASK: Bug diagnosis
-MODE: analysis
-CONTEXT: @**/*
-EXPECTED: Diagnosis, fix suggestions
-RULES: $(cat ~/.claude/workflows/cli-templates/prompts/analysis/01-diagnose-bug-root-cause.txt)
-" -m gpt-5 --skip-git-repo-check -s danger-full-access
-```
-
-## Output
-
-- **With session**: `.workflow/WFS-[id]/.chat/bug-diagnosis-[timestamp].md`
-- **No session**: `.workflow/.scratchpad/bug-diagnosis-[desc]-[timestamp].md`
-
-## Notes
-
-- Template: `~/.claude/workflows/cli-templates/prompts/analysis/01-diagnose-bug-root-cause.txt`
-- See `intelligent-tools-strategy.md` for detailed tool usage

.claude/commands/cli/mode/code-analysis.md

@@ -1,137 +0,0 @@
----
-name: code-analysis
-description: Deep code analysis and debugging using CLI tools with specialized template
-argument-hint: "[--agent] [--tool codex|gemini|qwen] [--enhance] [--cd path] analysis target"
-allowed-tools: SlashCommand(*), Bash(*), Task(*)
----
-
-# CLI Mode: Code Analysis (/cli:mode:code-analysis)
-
-## Purpose
-
-Systematic code analysis with execution path tracing template (`~/.claude/workflows/cli-templates/prompts/analysis/01-trace-code-execution.txt`).
-
-**Tool Selection**:
-- **gemini** (default) - Best for code analysis and tracing
-- **qwen** - Fallback when Gemini unavailable
-- **codex** - Alternative for complex analysis tasks
-
-**Key Feature**: `--cd` flag for directory-scoped analysis
-
-## Parameters
-
-- `--tool <gemini|qwen|codex>` - Tool selection (default: gemini)
-- `--agent` - Use cli-execution-agent for automated context discovery
-- `--enhance` - Enhance analysis target with `/enhance-prompt` first
-- `--cd "path"` - Target directory for focused analysis
-- `<analysis-target>` (Required) - Code analysis target or question
-
-## Tool Usage
-
-**Gemini** (Primary):
-```bash
-/cli:mode:code-analysis --tool gemini "trace auth flow"
-# OR (default)
-/cli:mode:code-analysis "trace auth flow"
-```
-
-**Qwen** (Fallback):
-```bash
-/cli:mode:code-analysis --tool qwen "trace auth flow"
-```
-
-**Codex** (Alternative):
-```bash
-/cli:mode:code-analysis --tool codex "trace auth flow"
-```
-
-## Execution Flow
-
-### Standard Mode (Default)
-
-1. Parse tool selection (default: gemini)
-2. Optional: enhance analysis target with `/enhance-prompt`
-3. Detect target directory from `--cd` or auto-infer
-4. Build command with template
-5. Execute analysis (read-only)
-6. Save to `.workflow/WFS-[id]/.chat/code-analysis-[timestamp].md`
-
-### Agent Mode (`--agent` flag)
-
-Delegates to `cli-execution-agent` for intelligent context discovery and analysis.
-
-## Core Rules
-
-- **Read-only**: Analyzes code, does NOT modify files
-- **Template**: Uses `~/.claude/workflows/cli-templates/prompts/analysis/01-trace-code-execution.txt` for systematic analysis
-- **Output**: Saves to `.workflow/WFS-[id]/.chat/`
-
-## CLI Command Templates
-
-**Gemini/Qwen** (default, read-only analysis):
-```bash
-cd [dir] && gemini -p "
-PURPOSE: [goal]
-TASK: Execution path tracing
-MODE: analysis
-CONTEXT: @**/*
-EXPECTED: Trace, call diagram
-RULES: $(cat ~/.claude/workflows/cli-templates/prompts/analysis/01-trace-code-execution.txt)
-"
-# Qwen: Replace 'gemini' with 'qwen'
-```
-
-**Codex** (analysis + optimization suggestions):
-```bash
-codex -C [dir] --full-auto exec "
-PURPOSE: [goal]
-TASK: Path analysis
-MODE: analysis
-CONTEXT: @**/*
-EXPECTED: Trace, optimization
-RULES: $(cat ~/.claude/workflows/cli-templates/prompts/analysis/01-trace-code-execution.txt)
-" -m gpt-5 --skip-git-repo-check -s danger-full-access
-```
-
-## Agent Execution Context
-
-When `--agent` flag is used, delegate to agent:
-
-```javascript
-Task(
-  subagent_type="cli-execution-agent",
-  description="Code execution path analysis",
-  prompt=`
-    Task: ${analysis_target}
-    Mode: code-analysis
-    Tool: ${tool_flag || 'auto-select'}  // gemini|qwen|codex
-    Directory: ${cd_path || 'auto-detect'}
-    Template: ~/.claude/workflows/cli-templates/prompts/analysis/01-trace-code-execution.txt
-
-    Agent responsibilities:
-    1. Context Discovery:
-       - Identify entry points and call chains
-       - Discover related files (MCP/ripgrep)
-       - Map execution flow paths
-
-    2. CLI Command Generation:
-       - Build Gemini/Qwen/Codex command
-       - Include discovered context
-       - Apply ~/.claude/workflows/cli-templates/prompts/analysis/01-trace-code-execution.txt template
-
-    3. Execution & Output:
-       - Execute analysis with selected tool
-       - Save to .workflow/WFS-[id]/.chat/
-  `
-)
-```
-
-## Output
-
-- **With session**: `.workflow/WFS-[id]/.chat/code-analysis-[timestamp].md`
-- **No session**: `.workflow/.scratchpad/code-analysis-[desc]-[timestamp].md`
-
-## Notes
-
-- Template: `~/.claude/workflows/cli-templates/prompts/analysis/01-trace-code-execution.txt`
-- See `intelligent-tools-strategy.md` for detailed tool usage

.claude/commands/cli/mode/plan.md

@@ -1,129 +0,0 @@
----
-name: plan
-description: Project planning and architecture analysis using CLI tools
-argument-hint: "[--agent] [--tool codex|gemini|qwen] [--enhance] [--cd path] topic"
-allowed-tools: SlashCommand(*), Bash(*), Task(*)
----
-
-# CLI Mode: Plan (/cli:mode:plan)
-
-## Purpose
-
-Strategic software architecture planning template (`~/.claude/workflows/cli-templates/prompts/planning/01-plan-architecture-design.txt`).
-
-**Tool Selection**:
-- **gemini** (default) - Best for architecture planning
-- **qwen** - Fallback when Gemini unavailable
-- **codex** - Alternative for implementation planning
-
-## Parameters
-
-- `--tool <gemini|qwen|codex>` - Tool selection (default: gemini)
-- `--agent` - Use cli-execution-agent for automated context discovery
-- `--enhance` - Enhance task with `/enhance-prompt`
-- `--cd "path"` - Target directory for focused planning
-- `<planning-task>` (Required) - Architecture planning task or modification requirements
-
-## Tool Usage
-
-**Gemini** (Primary):
-```bash
---tool gemini  # or omit (default)
-```
-
-**Qwen** (Fallback):
-```bash
---tool qwen
-```
-
-**Codex** (Alternative):
-```bash
---tool codex
-```
-
-## Execution Flow
-
-### Standard Mode
-1. Parse tool selection (default: gemini)
-2. Optional: enhance with `/enhance-prompt`
-3. Detect directory from `--cd` or auto-infer
-4. Build command with template
-5. Execute planning (read-only, no code generation)
-6. Save to `.workflow/WFS-[id]/.chat/`
-
-### Agent Mode (`--agent`)
-
-Delegates to agent for intelligent planning:
-
-```javascript
-Task(
-  subagent_type="cli-execution-agent",
-  description="Architecture modification planning",
-  prompt=`
-    Task: ${planning_task}
-    Mode: architecture-planning
-    Tool: ${tool_flag || 'auto-select'}  // gemini|qwen|codex
-    Directory: ${cd_path || 'auto-detect'}
-    Template: ~/.claude/workflows/cli-templates/prompts/planning/01-plan-architecture-design.txt
-
-    Agent responsibilities:
-    1. Context Discovery:
-       - Analyze current architecture
-       - Identify affected components
-       - Map dependencies and impacts
-
-    2. CLI Command Generation:
-       - Build Gemini/Qwen/Codex command
-       - Include architecture context
-       - Apply ~/.claude/workflows/cli-templates/prompts/planning/01-plan-architecture-design.txt template
-
-    3. Execution & Output:
-       - Execute strategic planning
-       - Generate modification plan
-       - Save to .workflow/.chat/
-  `
-)
-```
-
-## Core Rules
-
-- **Planning only**: Creates modification plans, does NOT generate code
-- **Template**: Uses `~/.claude/workflows/cli-templates/prompts/planning/01-plan-architecture-design.txt` for strategic planning
-- **Output**: Saves to `.workflow/WFS-[id]/.chat/`
-
-## CLI Command Templates
-
-**Gemini/Qwen** (default, planning only):
-```bash
-cd [dir] && gemini -p "
-PURPOSE: [goal]
-TASK: Architecture planning
-MODE: analysis
-CONTEXT: @**/*
-EXPECTED: Modification plan, impact analysis
-RULES: $(cat ~/.claude/workflows/cli-templates/prompts/planning/01-plan-architecture-design.txt)
-"
-# Qwen: Replace 'gemini' with 'qwen'
-```
-
-**Codex** (planning + implementation guidance):
-```bash
-codex -C [dir] --full-auto exec "
-PURPOSE: [goal]
-TASK: Architecture planning
-MODE: analysis
-CONTEXT: @**/*
-EXPECTED: Plan, implementation roadmap
-RULES: $(cat ~/.claude/workflows/cli-templates/prompts/planning/01-plan-architecture-design.txt)
-" -m gpt-5 --skip-git-repo-check -s danger-full-access
-```
-
-## Output
-
-- **With session**: `.workflow/WFS-[id]/.chat/plan-[timestamp].md`
-- **No session**: `.workflow/.scratchpad/plan-[desc]-[timestamp].md`
-
-## Notes
-
-- Template: `~/.claude/workflows/cli-templates/prompts/planning/01-plan-architecture-design.txt`
-- See `intelligent-tools-strategy.md` for detailed tool usage

.claude/commands/enhance-prompt.md

@@ -1,37 +1,22 @@
 ---
 name: enhance-prompt
-description: Context-aware prompt enhancement using session memory and codebase analysis
+description: Enhanced prompt transformation using session memory and intent analysis with --enhance flag detection
 argument-hint: "user input to enhance"
 ---
 
 ## Overview
 
-Systematically enhances user prompts by combining session memory context with codebase patterns, translating ambiguous requests into actionable specifications.
+Systematically enhances user prompts by leveraging session memory context and intent analysis, translating ambiguous requests into actionable specifications.
 
 ## Core Protocol
 
 **Enhancement Pipeline:**
-`Intent Translation` → `Context Integration` → `Gemini Analysis (if needed)` → `Structured Output`
+`Intent Translation` → `Context Integration` → `Structured Output`
 
 **Context Sources:**
 - Session memory (conversation history, previous analysis)
-- Codebase patterns (via Gemini when triggered)
 - Implicit technical requirements
-
-## Gemini Trigger Logic
-
-```pseudo
-FUNCTION should_use_gemini(user_prompt):
-  critical_keywords = ["refactor", "migrate", "redesign", "auth", "payment", "security"]
-
-  RETURN (
-    prompt_affects_multiple_modules(user_prompt, threshold=3) OR
-    any_keyword_in_prompt(critical_keywords, user_prompt)
-  )
-END
-```
-
-**Gemini Integration:** ~/.claude/workflows/intelligent-tools-strategy.md
+- User intent patterns
 
 ## Enhancement Rules
 
@@ -47,22 +32,18 @@ END
 
 ### Context Integration Strategy
 
-**Session Memory First:**
+**Session Memory:**
 - Reference recent conversation context
 - Reuse previously identified patterns
 - Build on established understanding
-
-**Codebase Analysis (via Gemini):**
-- Only when complexity requires it
-- Focus on integration points
-- Identify existing patterns
+- Infer technical requirements from discussion
 
 **Example:**
 ```bash
 # User: "add login"
 # Session Memory: Previous auth discussion, JWT mentioned
 # Inferred: JWT-based auth, integrate with existing session management
-# Gemini (if multi-module): Analyze AuthService patterns, middleware structure
+# Action: Implement JWT authentication with session persistence
 ```
 
 ## Output Structure
@@ -76,7 +57,7 @@ ATTENTION: [Critical constraints]
 
 ### Output Examples
 
-**Simple (no Gemini):**
+**Example 1:**
 ```bash
 # Input: "fix login button"
 INTENT: Debug non-functional login button
@@ -85,28 +66,28 @@ ACTION: Check event binding → verify state updates → test auth flow
 ATTENTION: Preserve existing OAuth integration
 ```
 
-**Complex (with Gemini):**
+**Example 2:**
 ```bash
 # Input: "refactor payment code"
 INTENT: Restructure payment module for maintainability
-CONTEXT: Session memory - PCI compliance requirements
-         Gemini - PaymentService → StripeAdapter pattern identified
-ACTION: Extract reusable validators → isolate payment gateway logic
+CONTEXT: Session memory - PCI compliance requirements, Stripe integration patterns
+ACTION: Extract reusable validators → isolate payment gateway logic → maintain adapter pattern
 ATTENTION: Zero behavior change, maintain PCI compliance, full test coverage
 ```
 
-## Automatic Triggers
+## Enhancement Triggers
 
 - Ambiguous language: "fix", "improve", "clean up"
-- Multi-module impact (>3 modules)
+- Vague requests requiring clarification
+- Complex technical requirements
 - Architecture changes
 - Critical systems: auth, payment, security
-- Complex refactoring
+- Multi-step refactoring
 
 ## Key Principles
 
-1. **Memory First**: Leverage session context before analysis
-2. **Minimal Gemini**: Only when complexity demands it
-3. **Context Reuse**: Build on previous understanding
-4. **Clear Output**: Structured, actionable specifications
+1. **Session Memory First**: Leverage conversation context and established understanding
+2. **Context Reuse**: Build on previous discussions and decisions
+3. **Clear Output**: Structured, actionable specifications
+4. **Intent Clarification**: Transform vague requests into specific technical goals
 5. **Avoid Duplication**: Reference existing context, don't repeat

.claude/commands/issue/discover.md

@@ -0,0 +1,427 @@
+---
+name: issue:discover
+description: Discover potential issues from multiple perspectives (bug, UX, test, quality, security, performance, maintainability, best-practices) using CLI explore. Supports Exa external research for security and best-practices perspectives.
+argument-hint: "<path-pattern> [--perspectives=bug,ux,...] [--external]"
+allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*), Task(*), AskUserQuestion(*), Glob(*), Grep(*)
+---
+
+# Issue Discovery Command
+
+## Quick Start
+
+```bash
+# Discover issues in specific module (interactive perspective selection)
+/issue:discover src/auth/**
+
+# Discover with specific perspectives
+/issue:discover src/payment/** --perspectives=bug,security,test
+
+# Discover with external research for all perspectives
+/issue:discover src/api/** --external
+
+# Discover in multiple modules
+/issue:discover src/auth/**,src/payment/**
+```
+
+**Discovery Scope**: Specified modules/files only
+**Output Directory**: `.workflow/issues/discoveries/{discovery-id}/`
+**Available Perspectives**: bug, ux, test, quality, security, performance, maintainability, best-practices
+**Exa Integration**: Auto-enabled for security and best-practices perspectives
+**CLI Tools**: Gemini → Qwen → Codex (fallback chain)
+
+## What & Why
+
+### Core Concept
+Multi-perspective issue discovery orchestrator that explores code from different angles to identify potential bugs, UX improvements, test gaps, and other actionable items. Unlike code review (which assesses existing code quality), discovery focuses on **finding opportunities for improvement and potential problems**.
+
+**vs Code Review**:
+- **Code Review** (`review-module-cycle`): Evaluates code quality against standards
+- **Issue Discovery** (`issue:discover`): Finds actionable issues, bugs, and improvement opportunities
+
+### Value Proposition
+1. **Proactive Issue Detection**: Find problems before they become bugs
+2. **Multi-Perspective Analysis**: Each perspective surfaces different types of issues
+3. **External Benchmarking**: Compare against industry best practices via Exa
+4. **Direct Issue Integration**: Discoveries can be exported to issue tracker
+5. **Dashboard Management**: View, filter, and export discoveries via CCW dashboard
+
+## How It Works
+
+### Execution Flow
+
+```
+Phase 1: Discovery & Initialization
+   └─ Parse target pattern, create session, initialize output structure
+
+Phase 2: Interactive Perspective Selection
+   └─ AskUserQuestion for perspective selection (or use --perspectives)
+
+Phase 3: Parallel Perspective Analysis
+   ├─ Launch N @cli-explore-agent instances (one per perspective)
+   ├─ Security & Best-Practices auto-trigger Exa research
+   ├─ Agent writes perspective JSON, returns summary
+   └─ Update discovery-progress.json
+
+Phase 4: Aggregation & Prioritization
+   ├─ Collect agent return summaries
+   ├─ Load perspective JSON files
+   ├─ Merge findings, deduplicate by file+line
+   └─ Calculate priority scores
+
+Phase 5: Issue Generation & Summary
+   ├─ Convert high-priority discoveries to issue format
+   ├─ Write to discovery-issues.jsonl
+   ├─ Generate single summary.md from agent returns
+   └─ Update discovery-state.json to complete
+```
+
+## Perspectives
+
+### Available Perspectives
+
+| Perspective | Focus | Categories | Exa |
+|-------------|-------|------------|-----|
+| **bug** | Potential Bugs | edge-case, null-check, resource-leak, race-condition, boundary, exception-handling | - |
+| **ux** | User Experience | error-message, loading-state, feedback, accessibility, interaction, consistency | - |
+| **test** | Test Coverage | missing-test, edge-case-test, integration-gap, coverage-hole, assertion-quality | - |
+| **quality** | Code Quality | complexity, duplication, naming, documentation, code-smell, readability | - |
+| **security** | Security Issues | injection, auth, encryption, input-validation, data-exposure, access-control | ✓ |
+| **performance** | Performance | n-plus-one, memory-usage, caching, algorithm, blocking-operation, resource | - |
+| **maintainability** | Maintainability | coupling, cohesion, tech-debt, extensibility, module-boundary, interface-design | - |
+| **best-practices** | Best Practices | convention, pattern, framework-usage, anti-pattern, industry-standard | ✓ |
+
+### Interactive Perspective Selection
+
+When no `--perspectives` flag is provided, the command uses AskUserQuestion:
+
+```javascript
+AskUserQuestion({
+  questions: [{
+    question: "Select primary discovery focus:",
+    header: "Focus",
+    multiSelect: false,
+    options: [
+      { label: "Bug + Test + Quality", description: "Quick scan: potential bugs, test gaps, code quality (Recommended)" },
+      { label: "Security + Performance", description: "System audit: security issues, performance bottlenecks" },
+      { label: "Maintainability + Best-practices", description: "Long-term health: coupling, tech debt, conventions" },
+      { label: "Full analysis", description: "All 7 perspectives (comprehensive, takes longer)" }
+    ]
+  }]
+})
+```
+
+**Recommended Combinations**:
+- Quick scan: bug, test, quality
+- Full analysis: all perspectives
+- Security audit: security, bug, quality
+
+## Core Responsibilities
+
+### Orchestrator
+
+**Phase 1: Discovery & Initialization**
+
+```javascript
+// Step 1: Parse target pattern and resolve files
+const resolvedFiles = await expandGlobPattern(targetPattern);
+if (resolvedFiles.length === 0) {
+  throw new Error(`No files matched pattern: ${targetPattern}`);
+}
+
+// Step 2: Generate discovery ID
+const discoveryId = `DSC-${formatDate(new Date(), 'YYYYMMDD-HHmmss')}`;
+
+// Step 3: Create output directory
+const outputDir = `.workflow/issues/discoveries/${discoveryId}`;
+await mkdir(outputDir, { recursive: true });
+await mkdir(`${outputDir}/perspectives`, { recursive: true });
+
+// Step 4: Initialize unified discovery state (merged state+progress)
+await writeJson(`${outputDir}/discovery-state.json`, {
+  discovery_id: discoveryId,
+  target_pattern: targetPattern,
+  phase: "initialization",
+  created_at: new Date().toISOString(),
+  updated_at: new Date().toISOString(),
+  target: { files_count: { total: resolvedFiles.length }, project: {} },
+  perspectives: [],  // filled after selection: [{name, status, findings}]
+  external_research: { enabled: false, completed: false },
+  results: { total_findings: 0, issues_generated: 0, priority_distribution: {} }
+});
+```
+
+**Phase 2: Perspective Selection**
+
+```javascript
+// Check for --perspectives flag
+let selectedPerspectives = [];
+
+if (args.perspectives) {
+  selectedPerspectives = args.perspectives.split(',').map(p => p.trim());
+} else {
+  // Interactive selection via AskUserQuestion
+  const response = await AskUserQuestion({...});
+  selectedPerspectives = parseSelectedPerspectives(response);
+}
+
+// Validate and update state
+await updateDiscoveryState(outputDir, {
+  'metadata.perspectives': selectedPerspectives,
+  phase: 'parallel'
+});
+```
+
+**Phase 3: Parallel Perspective Analysis**
+
+Launch N agents in parallel (one per selected perspective):
+
+```javascript
+// Launch agents in parallel - agents write JSON and return summary
+const agentPromises = selectedPerspectives.map(perspective =>
+  Task({
+    subagent_type: "cli-explore-agent",
+    run_in_background: false,
+    description: `Discover ${perspective} issues`,
+    prompt: buildPerspectivePrompt(perspective, discoveryId, resolvedFiles, outputDir)
+  })
+);
+
+// Wait for all agents - collect their return summaries
+const results = await Promise.all(agentPromises);
+// results contain agent summaries for final report
+```
+
+**Phase 4: Aggregation & Prioritization**
+
+```javascript
+// Load all perspective JSON files written by agents
+const allFindings = [];
+for (const perspective of selectedPerspectives) {
+  const jsonPath = `${outputDir}/perspectives/${perspective}.json`;
+  if (await fileExists(jsonPath)) {
+    const data = await readJson(jsonPath);
+    allFindings.push(...data.findings.map(f => ({ ...f, perspective })));
+  }
+}
+
+// Deduplicate and prioritize
+const prioritizedFindings = deduplicateAndPrioritize(allFindings);
+
+// Update unified state
+await updateDiscoveryState(outputDir, {
+  phase: 'aggregation',
+  'results.total_findings': prioritizedFindings.length,
+  'results.priority_distribution': countByPriority(prioritizedFindings)
+});
+```
+
+**Phase 5: Issue Generation & Summary**
+
+```javascript
+// Convert high-priority findings to issues
+const issueWorthy = prioritizedFindings.filter(f =>
+  f.priority === 'critical' || f.priority === 'high' || f.priority_score >= 0.7
+);
+
+// Write discovery-issues.jsonl
+await writeJsonl(`${outputDir}/discovery-issues.jsonl`, issues);
+
+// Generate single summary.md from agent return summaries
+// Orchestrator briefly summarizes what agents returned (NO detailed reports)
+await writeSummaryFromAgentReturns(outputDir, results, prioritizedFindings, issues);
+
+// Update final state
+await updateDiscoveryState(outputDir, {
+  phase: 'complete',
+  updated_at: new Date().toISOString(),
+  'results.issues_generated': issues.length
+});
+```
+
+### Output File Structure
+
+```
+.workflow/issues/discoveries/
+├── index.json                           # Discovery session index
+└── {discovery-id}/
+    ├── discovery-state.json             # Unified state (merged state+progress)
+    ├── perspectives/
+    │   └── {perspective}.json           # Per-perspective findings
+    ├── external-research.json           # Exa research results (if enabled)
+    ├── discovery-issues.jsonl           # Generated candidate issues
+    └── summary.md                       # Single summary (from agent returns)
+```
+
+### Schema References
+
+**External Schema Files** (agent MUST read and follow exactly):
+
+| Schema | Path | Purpose |
+|--------|------|---------|
+| **Discovery State** | `~/.claude/workflows/cli-templates/schemas/discovery-state-schema.json` | Session state machine |
+| **Discovery Finding** | `~/.claude/workflows/cli-templates/schemas/discovery-finding-schema.json` | Perspective analysis results |
+
+### Agent Invocation Template
+
+**Perspective Analysis Agent**:
+
+```javascript
+Task({
+  subagent_type: "cli-explore-agent",
+  run_in_background: false,
+  description: `Discover ${perspective} issues`,
+  prompt: `
+    ## Task Objective
+    Discover potential ${perspective} issues in specified module files.
+
+    ## Discovery Context
+    - Discovery ID: ${discoveryId}
+    - Perspective: ${perspective}
+    - Target Pattern: ${targetPattern}
+    - Resolved Files: ${resolvedFiles.length} files
+    - Output Directory: ${outputDir}
+
+    ## MANDATORY FIRST STEPS
+    1. Read discovery state: ${outputDir}/discovery-state.json
+    2. Read schema: ~/.claude/workflows/cli-templates/schemas/discovery-finding-schema.json
+    3. Analyze target files for ${perspective} concerns
+
+    ## Output Requirements
+
+    **1. Write JSON file**: ${outputDir}/perspectives/${perspective}.json
+    - Follow discovery-finding-schema.json exactly
+    - Each finding: id, title, priority, category, description, file, line, snippet, suggested_issue, confidence
+
+    **2. Return summary** (DO NOT write report file):
+    - Return a brief text summary of findings
+    - Include: total findings, priority breakdown, key issues
+    - This summary will be used by orchestrator for final report
+
+    ## Perspective-Specific Guidance
+    ${getPerspectiveGuidance(perspective)}
+
+    ## Success Criteria
+    - [ ] JSON written to ${outputDir}/perspectives/${perspective}.json
+    - [ ] Summary returned with findings count and key issues
+    - [ ] Each finding includes actionable suggested_issue
+    - [ ] Priority uses lowercase enum: critical/high/medium/low
+  `
+})
+```
+
+**Exa Research Agent** (for security and best-practices):
+
+```javascript
+Task({
+  subagent_type: "cli-explore-agent",
+  run_in_background: false,
+  description: `External research for ${perspective} via Exa`,
+  prompt: `
+    ## Task Objective
+    Research industry best practices for ${perspective} using Exa search
+
+    ## Research Steps
+    1. Read project tech stack: .workflow/project-tech.json
+    2. Use Exa to search for best practices
+    3. Synthesize findings relevant to this project
+
+    ## Output Requirements
+
+    **1. Write JSON file**: ${outputDir}/external-research.json
+    - Include sources, key_findings, gap_analysis, recommendations
+
+    **2. Return summary** (DO NOT write report file):
+    - Brief summary of external research findings
+    - Key recommendations for the project
+
+    ## Success Criteria
+    - [ ] JSON written to ${outputDir}/external-research.json
+    - [ ] Summary returned with key recommendations
+    - [ ] Findings are relevant to project's tech stack
+  `
+})
+```
+
+### Perspective Guidance Reference
+
+```javascript
+function getPerspectiveGuidance(perspective) {
+  const guidance = {
+    bug: `
+      Focus: Null checks, edge cases, resource leaks, race conditions, boundary conditions, exception handling
+      Priority: Critical=data corruption/crash, High=malfunction, Medium=edge case issues, Low=minor
+    `,
+    ux: `
+      Focus: Error messages, loading states, feedback, accessibility, interaction patterns, form validation
+      Priority: Critical=inaccessible, High=confusing, Medium=inconsistent, Low=cosmetic
+    `,
+    test: `
+      Focus: Missing unit tests, edge case coverage, integration gaps, assertion quality, test isolation
+      Priority: Critical=no security tests, High=no core logic tests, Medium=weak coverage, Low=minor gaps
+    `,
+    quality: `
+      Focus: Complexity, duplication, naming, documentation, code smells, readability
+      Priority: Critical=unmaintainable, High=significant issues, Medium=naming/docs, Low=minor refactoring
+    `,
+    security: `
+      Focus: Input validation, auth/authz, injection, XSS/CSRF, data exposure, access control
+      Priority: Critical=auth bypass/injection, High=missing authz, Medium=weak validation, Low=headers
+    `,
+    performance: `
+      Focus: N+1 queries, memory leaks, caching, algorithm efficiency, blocking operations
+      Priority: Critical=memory leaks, High=N+1/inefficient, Medium=missing cache, Low=minor optimization
+    `,
+    maintainability: `
+      Focus: Coupling, interface design, tech debt, extensibility, module boundaries, configuration
+      Priority: Critical=unrelated code changes, High=unclear boundaries, Medium=coupling, Low=refactoring
+    `,
+    'best-practices': `
+      Focus: Framework conventions, language patterns, anti-patterns, deprecated APIs, coding standards
+      Priority: Critical=anti-patterns causing bugs, High=convention violations, Medium=style, Low=cosmetic
+    `
+  };
+  return guidance[perspective] || 'General code discovery analysis';
+}
+```
+
+## Dashboard Integration
+
+### Viewing Discoveries
+
+Open CCW dashboard to manage discoveries:
+
+```bash
+ccw view
+```
+
+Navigate to **Issues > Discovery** to:
+- View all discovery sessions
+- Filter findings by perspective and priority
+- Preview finding details
+- Select and export findings as issues
+
+### Exporting to Issues
+
+From the dashboard, select findings and click "Export as Issues" to:
+1. Convert discoveries to standard issue format
+2. Append to `.workflow/issues/issues.jsonl`
+3. Set status to `registered`
+4. Continue with `/issue:plan` workflow
+
+## Related Commands
+
+```bash
+# After discovery, plan solutions for exported issues
+/issue:plan DSC-001,DSC-002,DSC-003
+
+# Or use interactive management
+/issue:manage
+```
+
+## Best Practices
+
+1. **Start Focused**: Begin with specific modules rather than entire codebase
+2. **Use Quick Scan First**: Start with bug, test, quality for fast results
+3. **Review Before Export**: Not all discoveries warrant issues - use dashboard to filter
+4. **Combine Perspectives**: Run related perspectives together (e.g., security + bug)
+5. **Enable Exa for New Tech**: When using unfamiliar frameworks, enable external research

.claude/commands/issue/execute.md

@@ -0,0 +1,294 @@
+---
+name: execute
+description: Execute queue with codex using DAG-based parallel orchestration (solution-level)
+argument-hint: "[--parallel <n>] [--executor codex|gemini|agent]"
+allowed-tools: TodoWrite(*), Bash(*), Read(*), AskUserQuestion(*)
+---
+
+# Issue Execute Command (/issue:execute)
+
+## Overview
+
+Minimal orchestrator that dispatches **solution IDs** to executors. Each executor receives a complete solution with all its tasks.
+
+**Design Principles:**
+- `queue dag` → returns parallel batches with solution IDs (S-1, S-2, ...)
+- `detail <id>` → READ-ONLY solution fetch (returns full solution with all tasks)
+- `done <id>` → update solution completion status
+- No race conditions: status changes only via `done`
+- **Executor handles all tasks within a solution sequentially**
+
+## Usage
+
+```bash
+/issue:execute [FLAGS]
+
+# Examples
+/issue:execute                    # Execute with default parallelism
+/issue:execute --parallel 4       # Execute up to 4 tasks in parallel
+/issue:execute --executor agent   # Use agent instead of codex
+
+# Flags
+--parallel <n>        Max parallel executors (default: 3)
+--executor <type>     Force executor: codex|gemini|agent (default: codex)
+--dry-run             Show DAG and batches without executing
+```
+
+## Execution Flow
+
+```
+Phase 1: Get DAG
+   └─ ccw issue queue dag → { parallel_batches: [["S-1","S-2"], ["S-3"]] }
+
+Phase 2: Dispatch Parallel Batch
+   ├─ For each solution ID in batch (parallel):
+   │   ├─ Executor calls: ccw issue detail <id>  (READ-ONLY)
+   │   ├─ Executor gets FULL SOLUTION with all tasks
+   │   ├─ Executor implements all tasks sequentially (T1 → T2 → T3)
+   │   ├─ Executor tests + commits per task
+   │   └─ Executor calls: ccw issue done <id>
+   └─ Wait for batch completion
+
+Phase 3: Next Batch
+   └─ ccw issue queue dag → check for newly-ready solutions
+```
+
+## Implementation
+
+### Phase 1: Get DAG
+
+```javascript
+// Get dependency graph and parallel batches
+const dagJson = Bash(`ccw issue queue dag`).trim();
+const dag = JSON.parse(dagJson);
+
+if (dag.error || dag.ready_count === 0) {
+  console.log(dag.error || 'No solutions ready for execution');
+  console.log('Use /issue:queue to form a queue first');
+  return;
+}
+
+console.log(`
+## Queue DAG (Solution-Level)
+
+- Total Solutions: ${dag.total}
+- Ready: ${dag.ready_count}
+- Completed: ${dag.completed_count}
+- Parallel in batch 1: ${dag.parallel_batches[0]?.length || 0}
+`);
+
+// Dry run mode
+if (flags.dryRun) {
+  console.log('### Parallel Batches:\n');
+  dag.parallel_batches.forEach((batch, i) => {
+    console.log(`Batch ${i + 1}: ${batch.join(', ')}`);
+  });
+  return;
+}
+```
+
+### Phase 2: Dispatch Parallel Batch
+
+```javascript
+const parallelLimit = flags.parallel || 3;
+const executor = flags.executor || 'codex';
+
+// Process first batch (all solutions can run in parallel)
+const batch = dag.parallel_batches[0] || [];
+
+// Initialize TodoWrite
+TodoWrite({
+  todos: batch.map(id => ({
+    content: `Execute solution ${id}`,
+    status: 'pending',
+    activeForm: `Executing solution ${id}`
+  }))
+});
+
+// Dispatch all in parallel (up to limit)
+const chunks = [];
+for (let i = 0; i < batch.length; i += parallelLimit) {
+  chunks.push(batch.slice(i, i + parallelLimit));
+}
+
+for (const chunk of chunks) {
+  console.log(`\n### Executing Solutions: ${chunk.join(', ')}`);
+
+  // Launch all in parallel
+  const executions = chunk.map(solutionId => {
+    updateTodo(solutionId, 'in_progress');
+    return dispatchExecutor(solutionId, executor);
+  });
+
+  await Promise.all(executions);
+  chunk.forEach(id => updateTodo(id, 'completed'));
+}
+```
+
+### Executor Dispatch
+
+```javascript
+function dispatchExecutor(solutionId, executorType) {
+  // Executor fetches FULL SOLUTION via READ-ONLY detail command
+  // Executor handles all tasks within solution sequentially
+  // Then reports completion via done command
+  const prompt = `
+## Execute Solution ${solutionId}
+
+### Step 1: Get Solution (read-only)
+\`\`\`bash
+ccw issue detail ${solutionId}
+\`\`\`
+
+### Step 2: Execute All Tasks Sequentially
+The detail command returns a FULL SOLUTION with all tasks.
+Execute each task in order (T1 → T2 → T3 → ...):
+
+For each task:
+1. Follow task.implementation steps
+2. Run task.test commands
+3. Verify task.acceptance criteria
+4. Commit using task.commit specification
+
+### Step 3: Report Completion
+When ALL tasks in solution are done:
+\`\`\`bash
+ccw issue done ${solutionId} --result '{"summary": "...", "files_modified": [...], "tasks_completed": N}'
+\`\`\`
+
+If any task failed:
+\`\`\`bash
+ccw issue done ${solutionId} --fail --reason "Task TX failed: ..."
+\`\`\`
+`;
+
+  if (executorType === 'codex') {
+    return Bash(
+      `ccw cli -p "${escapePrompt(prompt)}" --tool codex --mode write --id exec-${solutionId}`,
+      { timeout: 7200000, run_in_background: true }  // 2hr for full solution
+    );
+  } else if (executorType === 'gemini') {
+    return Bash(
+      `ccw cli -p "${escapePrompt(prompt)}" --tool gemini --mode write --id exec-${solutionId}`,
+      { timeout: 3600000, run_in_background: true }
+    );
+  } else {
+    return Task({
+      subagent_type: 'code-developer',
+      run_in_background: false,
+      description: `Execute solution ${solutionId}`,
+      prompt: prompt
+    });
+  }
+}
+```
+
+### Phase 3: Check Next Batch
+
+```javascript
+// Refresh DAG after batch completes
+const refreshedDag = JSON.parse(Bash(`ccw issue queue dag`).trim());
+
+console.log(`
+## Batch Complete
+
+- Solutions Completed: ${refreshedDag.completed_count}/${refreshedDag.total}
+- Next ready: ${refreshedDag.ready_count}
+`);
+
+if (refreshedDag.ready_count > 0) {
+  console.log('Run `/issue:execute` again for next batch.');
+}
+```
+
+## Parallel Execution Model
+
+```
+┌─────────────────────────────────────────────────────────────┐
+│ Orchestrator                                                │
+├─────────────────────────────────────────────────────────────┤
+│ 1. ccw issue queue dag                                      │
+│    → { parallel_batches: [["S-1","S-2"], ["S-3"]] }        │
+│                                                             │
+│ 2. Dispatch batch 1 (parallel):                            │
+│    ┌──────────────────────┐ ┌──────────────────────┐       │
+│    │ Executor 1           │ │ Executor 2           │       │
+│    │ detail S-1           │ │ detail S-2           │       │
+│    │ → gets full solution │ │ → gets full solution │       │
+│    │ [T1→T2→T3 sequential]│ │ [T1→T2 sequential]   │       │
+│    │ done S-1             │ │ done S-2             │       │
+│    └──────────────────────┘ └──────────────────────┘       │
+│                                                             │
+│ 3. ccw issue queue dag (refresh)                           │
+│    → S-3 now ready (S-1 completed, file conflict resolved) │
+└─────────────────────────────────────────────────────────────┘
+```
+
+**Why this works for parallel:**
+- `detail <id>` is READ-ONLY → no race conditions
+- Each executor handles **all tasks within a solution** sequentially
+- `done <id>` updates only its own solution status
+- `queue dag` recalculates ready solutions after each batch
+- Solutions in same batch have NO file conflicts
+
+## CLI Endpoint Contract
+
+### `ccw issue queue dag`
+Returns dependency graph with parallel batches (solution-level):
+```json
+{
+  "queue_id": "QUE-...",
+  "total": 3,
+  "ready_count": 2,
+  "completed_count": 0,
+  "nodes": [
+    { "id": "S-1", "issue_id": "ISS-xxx", "status": "pending", "ready": true, "task_count": 3 },
+    { "id": "S-2", "issue_id": "ISS-yyy", "status": "pending", "ready": true, "task_count": 2 },
+    { "id": "S-3", "issue_id": "ISS-zzz", "status": "pending", "ready": false, "depends_on": ["S-1"] }
+  ],
+  "parallel_batches": [["S-1", "S-2"], ["S-3"]]
+}
+```
+
+### `ccw issue detail <item_id>`
+Returns FULL SOLUTION with all tasks (READ-ONLY):
+```json
+{
+  "item_id": "S-1",
+  "issue_id": "ISS-xxx",
+  "solution_id": "SOL-xxx",
+  "status": "pending",
+  "solution": {
+    "id": "SOL-xxx",
+    "approach": "...",
+    "tasks": [
+      { "id": "T1", "title": "...", "implementation": [...], "test": {...} },
+      { "id": "T2", "title": "...", "implementation": [...], "test": {...} },
+      { "id": "T3", "title": "...", "implementation": [...], "test": {...} }
+    ],
+    "exploration_context": { "relevant_files": [...] }
+  },
+  "execution_hints": { "executor": "codex", "estimated_minutes": 180 }
+}
+```
+
+### `ccw issue done <item_id>`
+Marks solution completed/failed, updates queue state, checks for queue completion.
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| No queue | Run /issue:queue first |
+| No ready solutions | Dependencies blocked, check DAG |
+| Executor timeout | Solution not marked done, can retry |
+| Solution failure | Use `ccw issue retry` to reset |
+| Partial task failure | Executor reports which task failed via `done --fail` |
+
+## Related Commands
+
+- `/issue:plan` - Plan issues with solutions
+- `/issue:queue` - Form execution queue
+- `ccw issue queue dag` - View dependency graph
+- `ccw issue detail <id>` - View task details
+- `ccw issue retry` - Reset failed tasks

.claude/commands/issue/manage.md

@@ -0,0 +1,113 @@
+---
+name: manage
+description: Interactive issue management (CRUD) via ccw cli endpoints with menu-driven interface
+argument-hint: "[issue-id] [--action list|view|edit|delete|bulk]"
+allowed-tools: TodoWrite(*), Bash(*), Read(*), Write(*), AskUserQuestion(*), Task(*)
+---
+
+# Issue Manage Command (/issue:manage)
+
+## Overview
+
+Interactive menu-driven interface for issue management using `ccw issue` CLI endpoints:
+- **List**: Browse and filter issues
+- **View**: Detailed issue inspection
+- **Edit**: Modify issue fields
+- **Delete**: Remove issues
+- **Bulk**: Batch operations on multiple issues
+
+## CLI Endpoints Reference
+
+```bash
+# Core endpoints (ccw issue)
+ccw issue list                      # List all issues
+ccw issue list <id> --json          # Get issue details
+ccw issue status <id>               # Detailed status
+ccw issue init <id> --title "..."   # Create issue
+ccw issue task <id> --title "..."   # Add task
+ccw issue bind <id> <solution-id>   # Bind solution
+
+# Queue management
+ccw issue queue                     # List current queue
+ccw issue queue add <id>            # Add to queue
+ccw issue queue list                # Queue history
+ccw issue queue switch <queue-id>   # Switch queue
+ccw issue queue archive             # Archive queue
+ccw issue queue delete <queue-id>   # Delete queue
+ccw issue next                      # Get next task
+ccw issue done <queue-id>           # Mark completed
+ccw issue complete <item-id>        # (legacy alias for done)
+```
+
+## Usage
+
+```bash
+# Interactive mode (menu-driven)
+/issue:manage
+
+# Direct to specific issue
+/issue:manage GH-123
+
+# Direct action
+/issue:manage --action list
+/issue:manage GH-123 --action edit
+```
+
+## Implementation
+
+This command delegates to the `issue-manage` skill for detailed implementation.
+
+### Entry Point
+
+```javascript
+const issueId = parseIssueId(userInput);
+const action = flags.action;
+
+// Show main menu if no action specified
+if (!action) {
+  await showMainMenu(issueId);
+} else {
+  await executeAction(action, issueId);
+}
+```
+
+### Main Menu Flow
+
+1. **Dashboard**: Fetch issues summary via `ccw issue list --json`
+2. **Menu**: Present action options via AskUserQuestion
+3. **Route**: Execute selected action (List/View/Edit/Delete/Bulk)
+4. **Loop**: Return to menu after each action
+
+### Available Actions
+
+| Action | Description | CLI Command |
+|--------|-------------|-------------|
+| List | Browse with filters | `ccw issue list --json` |
+| View | Detail view | `ccw issue status <id> --json` |
+| Edit | Modify fields | Update `issues.jsonl` |
+| Delete | Remove issue | Clean up all related files |
+| Bulk | Batch operations | Multi-select + batch update |
+
+## Data Files
+
+| File | Purpose |
+|------|---------|
+| `.workflow/issues/issues.jsonl` | Issue records |
+| `.workflow/issues/solutions/<id>.jsonl` | Solutions per issue |
+| `.workflow/issues/queue.json` | Execution queue |
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| No issues found | Suggest creating with /issue:new |
+| Issue not found | Show available issues, ask for correction |
+| Invalid selection | Show error, re-prompt |
+| Write failure | Check permissions, show error |
+
+## Related Commands
+
+- `/issue:new` - Create structured issue
+- `/issue:plan` - Plan solution for issue
+- `/issue:queue` - Form execution queue
+- `/issue:execute` - Execute queued tasks

.claude/commands/issue/new.md

@@ -0,0 +1,463 @@
+---
+name: new
+description: Create structured issue from GitHub URL or text description, extracting key elements into issues.jsonl
+argument-hint: "<github-url | text-description> [--priority 1-5] [--labels label1,label2]"
+allowed-tools: TodoWrite(*), Bash(*), Read(*), Write(*), WebFetch(*), AskUserQuestion(*)
+---
+
+# Issue New Command (/issue:new)
+
+## Overview
+
+Creates a new structured issue from either:
+1. **GitHub Issue URL** - Fetches and parses issue content via `gh` CLI
+2. **Text Description** - Parses natural language into structured fields
+
+Outputs a well-formed issue entry to `.workflow/issues/issues.jsonl`.
+
+## Issue Structure (Closed-Loop)
+
+```typescript
+interface Issue {
+  id: string;                    // GH-123 or ISS-YYYYMMDD-HHMMSS
+  title: string;                 // Issue title (clear, concise)
+  status: 'registered';          // Initial status
+  priority: number;              // 1 (critical) to 5 (low)
+  context: string;               // Problem description
+  source: 'github' | 'text' | 'discovery';  // Input source type
+  source_url?: string;           // GitHub URL if applicable
+  labels?: string[];             // Categorization labels
+
+  // Structured extraction
+  problem_statement: string;     // What is the problem?
+  expected_behavior?: string;    // What should happen?
+  actual_behavior?: string;      // What actually happens?
+  affected_components?: string[];// Files/modules affected
+  reproduction_steps?: string[]; // Steps to reproduce
+
+  // Discovery context (when source='discovery')
+  discovery_context?: {
+    discovery_id: string;        // Source discovery session
+    perspective: string;         // bug, test, quality, etc.
+    category: string;            // Finding category
+    file: string;                // Primary affected file
+    line: number;                // Line number
+    snippet?: string;            // Code snippet
+    confidence: number;          // Agent confidence (0-1)
+    suggested_fix?: string;      // Suggested remediation
+  };
+
+  // Closed-loop requirements (guide plan generation)
+  lifecycle_requirements: {
+    test_strategy: 'unit' | 'integration' | 'e2e' | 'manual' | 'auto';
+    regression_scope: 'affected' | 'related' | 'full';  // Which tests to run
+    acceptance_type: 'automated' | 'manual' | 'both';   // How to verify
+    commit_strategy: 'per-task' | 'squash' | 'atomic';  // Commit granularity
+  };
+
+  // Metadata
+  bound_solution_id: null;
+  solution_count: 0;
+  created_at: string;
+  updated_at: string;
+}
+```
+
+## Lifecycle Requirements
+
+The `lifecycle_requirements` field guides downstream commands (`/issue:plan`, `/issue:execute`):
+
+| Field | Options | Purpose |
+|-------|---------|---------|
+| `test_strategy` | `unit`, `integration`, `e2e`, `manual`, `auto` | Which test types to generate |
+| `regression_scope` | `affected`, `related`, `full` | Which tests to run for regression |
+| `acceptance_type` | `automated`, `manual`, `both` | How to verify completion |
+| `commit_strategy` | `per-task`, `squash`, `atomic` | Commit granularity |
+
+> **Note**: Task structure (SolutionTask) is defined in `/issue:plan` - see `.claude/commands/issue/plan.md`
+
+## Usage
+
+```bash
+# From GitHub URL
+/issue:new https://github.com/owner/repo/issues/123
+
+# From text description
+/issue:new "Login fails when password contains special characters. Expected: successful login. Actual: 500 error. Affects src/auth/*"
+
+# With options
+/issue:new <url-or-text> --priority 2 --labels "bug,auth"
+```
+
+## Implementation
+
+### Phase 1: Input Detection
+
+```javascript
+const input = userInput.trim();
+const flags = parseFlags(userInput);  // --priority, --labels
+
+// Detect input type
+const isGitHubUrl = input.match(/github\.com\/[\w-]+\/[\w-]+\/issues\/\d+/);
+const isGitHubShort = input.match(/^#(\d+)$/);  // #123 format
+
+let issueData = {};
+
+if (isGitHubUrl || isGitHubShort) {
+  // GitHub issue - fetch via gh CLI
+  issueData = await fetchGitHubIssue(input);
+} else {
+  // Text description - parse structure
+  issueData = await parseTextDescription(input);
+}
+```
+
+### Phase 2: GitHub Issue Fetching
+
+```javascript
+async function fetchGitHubIssue(urlOrNumber) {
+  let issueRef;
+  
+  if (urlOrNumber.startsWith('http')) {
+    // Extract owner/repo/number from URL
+    const match = urlOrNumber.match(/github\.com\/([\w-]+)\/([\w-]+)\/issues\/(\d+)/);
+    if (!match) throw new Error('Invalid GitHub URL');
+    issueRef = `${match[1]}/${match[2]}#${match[3]}`;
+  } else {
+    // #123 format - use current repo
+    issueRef = urlOrNumber.replace('#', '');
+  }
+  
+  // Fetch via gh CLI
+  const result = Bash(`gh issue view ${issueRef} --json number,title,body,labels,state,url`);
+  const ghIssue = JSON.parse(result);
+  
+  // Parse body for structure
+  const parsed = parseIssueBody(ghIssue.body);
+  
+  return {
+    id: `GH-${ghIssue.number}`,
+    title: ghIssue.title,
+    source: 'github',
+    source_url: ghIssue.url,
+    labels: ghIssue.labels.map(l => l.name),
+    context: ghIssue.body,
+    ...parsed
+  };
+}
+
+function parseIssueBody(body) {
+  // Extract structured sections from markdown body
+  const sections = {};
+  
+  // Problem/Description
+  const problemMatch = body.match(/##?\s*(problem|description|issue)[:\s]*([\s\S]*?)(?=##|$)/i);
+  if (problemMatch) sections.problem_statement = problemMatch[2].trim();
+  
+  // Expected behavior
+  const expectedMatch = body.match(/##?\s*(expected|should)[:\s]*([\s\S]*?)(?=##|$)/i);
+  if (expectedMatch) sections.expected_behavior = expectedMatch[2].trim();
+  
+  // Actual behavior
+  const actualMatch = body.match(/##?\s*(actual|current)[:\s]*([\s\S]*?)(?=##|$)/i);
+  if (actualMatch) sections.actual_behavior = actualMatch[2].trim();
+  
+  // Steps to reproduce
+  const stepsMatch = body.match(/##?\s*(steps|reproduce)[:\s]*([\s\S]*?)(?=##|$)/i);
+  if (stepsMatch) {
+    const stepsText = stepsMatch[2].trim();
+    sections.reproduction_steps = stepsText
+      .split('\n')
+      .filter(line => line.match(/^\s*[\d\-\*]/))
+      .map(line => line.replace(/^\s*[\d\.\-\*]\s*/, '').trim());
+  }
+  
+  // Affected components (from file references)
+  const fileMatches = body.match(/`[^`]*\.(ts|js|tsx|jsx|py|go|rs)[^`]*`/g);
+  if (fileMatches) {
+    sections.affected_components = [...new Set(fileMatches.map(f => f.replace(/`/g, '')))];
+  }
+  
+  // Fallback: use entire body as problem statement
+  if (!sections.problem_statement) {
+    sections.problem_statement = body.substring(0, 500);
+  }
+  
+  return sections;
+}
+```
+
+### Phase 3: Text Description Parsing
+
+```javascript
+async function parseTextDescription(text) {
+  // Generate unique ID
+  const id = `ISS-${new Date().toISOString().replace(/[-:T]/g, '').slice(0, 14)}`;
+  
+  // Extract structured elements using patterns
+  const result = {
+    id,
+    source: 'text',
+    title: '',
+    problem_statement: '',
+    expected_behavior: null,
+    actual_behavior: null,
+    affected_components: [],
+    reproduction_steps: []
+  };
+  
+  // Pattern: "Title. Description. Expected: X. Actual: Y. Affects: files"
+  const sentences = text.split(/\.(?=\s|$)/);
+  
+  // First sentence as title
+  result.title = sentences[0]?.trim() || 'Untitled Issue';
+  
+  // Look for keywords
+  for (const sentence of sentences) {
+    const s = sentence.trim();
+    
+    if (s.match(/^expected:?\s*/i)) {
+      result.expected_behavior = s.replace(/^expected:?\s*/i, '');
+    } else if (s.match(/^actual:?\s*/i)) {
+      result.actual_behavior = s.replace(/^actual:?\s*/i, '');
+    } else if (s.match(/^affects?:?\s*/i)) {
+      const components = s.replace(/^affects?:?\s*/i, '').split(/[,\s]+/);
+      result.affected_components = components.filter(c => c.includes('/') || c.includes('.'));
+    } else if (s.match(/^steps?:?\s*/i)) {
+      result.reproduction_steps = s.replace(/^steps?:?\s*/i, '').split(/[,;]/);
+    } else if (!result.problem_statement && s.length > 10) {
+      result.problem_statement = s;
+    }
+  }
+  
+  // Fallback problem statement
+  if (!result.problem_statement) {
+    result.problem_statement = text.substring(0, 300);
+  }
+  
+  return result;
+}
+```
+
+### Phase 4: Lifecycle Configuration
+
+```javascript
+// Ask for lifecycle requirements (or use smart defaults)
+const lifecycleAnswer = AskUserQuestion({
+  questions: [
+    {
+      question: 'Test strategy for this issue?',
+      header: 'Test',
+      multiSelect: false,
+      options: [
+        { label: 'auto', description: 'Auto-detect based on affected files (Recommended)' },
+        { label: 'unit', description: 'Unit tests only' },
+        { label: 'integration', description: 'Integration tests' },
+        { label: 'e2e', description: 'End-to-end tests' },
+        { label: 'manual', description: 'Manual testing only' }
+      ]
+    },
+    {
+      question: 'Regression scope?',
+      header: 'Regression',
+      multiSelect: false,
+      options: [
+        { label: 'affected', description: 'Only affected module tests (Recommended)' },
+        { label: 'related', description: 'Affected + dependent modules' },
+        { label: 'full', description: 'Full test suite' }
+      ]
+    },
+    {
+      question: 'Commit strategy?',
+      header: 'Commit',
+      multiSelect: false,
+      options: [
+        { label: 'per-task', description: 'One commit per task (Recommended)' },
+        { label: 'atomic', description: 'Single commit for entire issue' },
+        { label: 'squash', description: 'Squash at the end' }
+      ]
+    }
+  ]
+});
+
+const lifecycle = {
+  test_strategy: lifecycleAnswer.test || 'auto',
+  regression_scope: lifecycleAnswer.regression || 'affected',
+  acceptance_type: 'automated',
+  commit_strategy: lifecycleAnswer.commit || 'per-task'
+};
+
+issueData.lifecycle_requirements = lifecycle;
+```
+
+### Phase 5: User Confirmation
+
+```javascript
+// Show parsed data and ask for confirmation
+console.log(`
+## Parsed Issue
+
+**ID**: ${issueData.id}
+**Title**: ${issueData.title}
+**Source**: ${issueData.source}${issueData.source_url ? ` (${issueData.source_url})` : ''}
+
+### Problem Statement
+${issueData.problem_statement}
+
+${issueData.expected_behavior ? `### Expected Behavior\n${issueData.expected_behavior}\n` : ''}
+${issueData.actual_behavior ? `### Actual Behavior\n${issueData.actual_behavior}\n` : ''}
+${issueData.affected_components?.length ? `### Affected Components\n${issueData.affected_components.map(c => `- ${c}`).join('\n')}\n` : ''}
+${issueData.reproduction_steps?.length ? `### Reproduction Steps\n${issueData.reproduction_steps.map((s, i) => `${i+1}. ${s}`).join('\n')}\n` : ''}
+
+### Lifecycle Configuration
+- **Test Strategy**: ${lifecycle.test_strategy}
+- **Regression Scope**: ${lifecycle.regression_scope}
+- **Commit Strategy**: ${lifecycle.commit_strategy}
+`);
+
+// Ask user to confirm or edit
+const answer = AskUserQuestion({
+  questions: [{
+    question: 'Create this issue?',
+    header: 'Confirm',
+    multiSelect: false,
+    options: [
+      { label: 'Create', description: 'Save issue to issues.jsonl' },
+      { label: 'Edit Title', description: 'Modify the issue title' },
+      { label: 'Edit Priority', description: 'Change priority (1-5)' },
+      { label: 'Cancel', description: 'Discard and exit' }
+    ]
+  }]
+});
+
+if (answer.includes('Cancel')) {
+  console.log('Issue creation cancelled.');
+  return;
+}
+
+if (answer.includes('Edit Title')) {
+  const titleAnswer = AskUserQuestion({
+    questions: [{
+      question: 'Enter new title:',
+      header: 'Title',
+      multiSelect: false,
+      options: [
+        { label: issueData.title.substring(0, 40), description: 'Keep current' }
+      ]
+    }]
+  });
+  // Handle custom input via "Other"
+  if (titleAnswer.customText) {
+    issueData.title = titleAnswer.customText;
+  }
+}
+```
+
+### Phase 6: Write to JSONL
+
+```javascript
+// Construct final issue object
+const priority = flags.priority ? parseInt(flags.priority) : 3;
+const labels = flags.labels ? flags.labels.split(',').map(l => l.trim()) : [];
+
+const newIssue = {
+  id: issueData.id,
+  title: issueData.title,
+  status: 'registered',
+  priority,
+  context: issueData.problem_statement,
+  source: issueData.source,
+  source_url: issueData.source_url || null,
+  labels: [...(issueData.labels || []), ...labels],
+
+  // Structured fields
+  problem_statement: issueData.problem_statement,
+  expected_behavior: issueData.expected_behavior || null,
+  actual_behavior: issueData.actual_behavior || null,
+  affected_components: issueData.affected_components || [],
+  reproduction_steps: issueData.reproduction_steps || [],
+
+  // Closed-loop lifecycle requirements
+  lifecycle_requirements: issueData.lifecycle_requirements || {
+    test_strategy: 'auto',
+    regression_scope: 'affected',
+    acceptance_type: 'automated',
+    commit_strategy: 'per-task'
+  },
+
+  // Metadata
+  bound_solution_id: null,
+  solution_count: 0,
+  created_at: new Date().toISOString(),
+  updated_at: new Date().toISOString()
+};
+
+// Ensure directory exists
+Bash('mkdir -p .workflow/issues');
+
+// Append to issues.jsonl
+const issuesPath = '.workflow/issues/issues.jsonl';
+Bash(`echo '${JSON.stringify(newIssue)}' >> "${issuesPath}"`);
+
+console.log(`
+## Issue Created
+
+**ID**: ${newIssue.id}
+**Title**: ${newIssue.title}
+**Priority**: ${newIssue.priority}
+**Labels**: ${newIssue.labels.join(', ') || 'none'}
+**Source**: ${newIssue.source}
+
+### Next Steps
+1. Plan solution: \`/issue:plan ${newIssue.id}\`
+2. View details: \`ccw issue status ${newIssue.id}\`
+3. Manage issues: \`/issue:manage\`
+`);
+```
+
+## Examples
+
+### GitHub Issue
+
+```bash
+/issue:new https://github.com/myorg/myrepo/issues/42 --priority 2
+
+# Output:
+## Issue Created
+**ID**: GH-42
+**Title**: Fix memory leak in WebSocket handler
+**Priority**: 2
+**Labels**: bug, performance
+**Source**: github (https://github.com/myorg/myrepo/issues/42)
+```
+
+### Text Description
+
+```bash
+/issue:new "API rate limiting not working. Expected: 429 after 100 requests. Actual: No limit. Affects src/middleware/rate-limit.ts"
+
+# Output:
+## Issue Created
+**ID**: ISS-20251227-142530
+**Title**: API rate limiting not working
+**Priority**: 3
+**Labels**: none
+**Source**: text
+```
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| Invalid GitHub URL | Show format hint, ask for correction |
+| gh CLI not available | Fall back to WebFetch for public issues |
+| Empty description | Prompt user for required fields |
+| Duplicate issue ID | Auto-increment or suggest merge |
+| Parse failure | Show raw input, ask for manual structuring |
+
+## Related Commands
+
+- `/issue:plan` - Plan solution for issue
+- `/issue:manage` - Interactive issue management
+- `ccw issue list` - List all issues
+- `ccw issue status <id>` - View issue details

.claude/commands/issue/plan.md

@@ -0,0 +1,327 @@
+---
+name: plan
+description: Batch plan issue resolution using issue-plan-agent (explore + plan closed-loop)
+argument-hint: "--all-pending <issue-id>[,<issue-id>,...] [--batch-size 3] "
+allowed-tools: TodoWrite(*), Task(*), SlashCommand(*), AskUserQuestion(*), Bash(*), Read(*), Write(*)
+---
+
+# Issue Plan Command (/issue:plan)
+
+## Overview
+
+Unified planning command using **issue-plan-agent** that combines exploration and planning into a single closed-loop workflow.
+
+## Output Requirements
+
+**Generate Files:**
+1. `.workflow/issues/solutions/{issue-id}.jsonl` - Solution with tasks for each issue
+
+**Return Summary:**
+```json
+{
+  "bound": [{ "issue_id": "...", "solution_id": "...", "task_count": N }],
+  "pending_selection": [{ "issue_id": "...", "solutions": [...] }],
+  "conflicts": [{ "file": "...", "issues": [...] }]
+}
+```
+
+**Completion Criteria:**
+- [ ] Solution file generated for each issue
+- [ ] Single solution → auto-bound via `ccw issue bind`
+- [ ] Multiple solutions → returned for user selection
+- [ ] Tasks conform to schema: `cat .claude/workflows/cli-templates/schemas/solution-schema.json`
+- [ ] Each task has quantified `acceptance.criteria`
+
+## Core Capabilities
+
+- **Closed-loop agent**: issue-plan-agent combines explore + plan
+- Batch processing: 1 agent processes 1-3 issues
+- ACE semantic search integrated into planning
+- Solution with executable tasks and delivery criteria
+- Automatic solution registration and binding
+
+## Storage Structure (Flat JSONL)
+
+```
+.workflow/issues/
+├── issues.jsonl              # All issues (one per line)
+├── queue.json                # Execution queue
+└── solutions/
+    ├── {issue-id}.jsonl      # Solutions for issue (one per line)
+    └── ...
+```
+
+## Usage
+
+```bash
+/issue:plan <issue-id>[,<issue-id>,...] [FLAGS]
+
+# Examples
+/issue:plan GH-123                    # Single issue
+/issue:plan GH-123,GH-124,GH-125      # Batch (up to 3)
+/issue:plan --all-pending             # All pending issues
+
+# Flags
+--batch-size <n>      Max issues per agent batch (default: 3)
+```
+
+## Execution Process
+
+```
+Phase 1: Issue Loading
+   ├─ Parse input (single, comma-separated, or --all-pending)
+   ├─ Fetch issue metadata (ID, title, tags)
+   ├─ Validate issues exist (create if needed)
+   └─ Group by similarity (shared tags or title keywords, max 3 per batch)
+
+Phase 2: Unified Explore + Plan (issue-plan-agent)
+   ├─ Launch issue-plan-agent per batch
+   ├─ Agent performs:
+   │   ├─ ACE semantic search for each issue
+   │   ├─ Codebase exploration (files, patterns, dependencies)
+   │   ├─ Solution generation with task breakdown
+   │   └─ Conflict detection across issues
+   └─ Output: solution JSON per issue
+
+Phase 3: Solution Registration & Binding
+   ├─ Append solutions to solutions/{issue-id}.jsonl
+   ├─ Single solution per issue → auto-bind
+   ├─ Multiple candidates → AskUserQuestion to select
+   └─ Update issues.jsonl with bound_solution_id
+
+Phase 4: Summary
+   ├─ Display bound solutions
+   ├─ Show task counts per issue
+   └─ Display next steps (/issue:queue)
+```
+
+## Implementation
+
+### Phase 1: Issue Loading (ID + Title + Tags)
+
+```javascript
+const batchSize = flags.batchSize || 3;
+let issues = [];  // {id, title, tags}
+
+if (flags.allPending) {
+  // Get pending issues with metadata via CLI (JSON output)
+  const result = Bash(`ccw issue list --status pending,registered --json`).trim();
+  const parsed = result ? JSON.parse(result) : [];
+  issues = parsed.map(i => ({ id: i.id, title: i.title || '', tags: i.tags || [] }));
+
+  if (issues.length === 0) {
+    console.log('No pending issues found.');
+    return;
+  }
+  console.log(`Found ${issues.length} pending issues`);
+} else {
+  // Parse comma-separated issue IDs, fetch metadata
+  const ids = userInput.includes(',')
+    ? userInput.split(',').map(s => s.trim())
+    : [userInput.trim()];
+
+  for (const id of ids) {
+    Bash(`ccw issue init ${id} --title "Issue ${id}" 2>/dev/null || true`);
+    const info = Bash(`ccw issue status ${id} --json`).trim();
+    const parsed = info ? JSON.parse(info) : {};
+    issues.push({ id, title: parsed.title || '', tags: parsed.tags || [] });
+  }
+}
+
+// Intelligent grouping by similarity (tags → title keywords)
+function groupBySimilarity(issues, maxSize) {
+  const batches = [];
+  const used = new Set();
+
+  for (const issue of issues) {
+    if (used.has(issue.id)) continue;
+
+    const batch = [issue];
+    used.add(issue.id);
+    const issueTags = new Set(issue.tags);
+    const issueWords = new Set(issue.title.toLowerCase().split(/\s+/));
+
+    // Find similar issues
+    for (const other of issues) {
+      if (used.has(other.id) || batch.length >= maxSize) continue;
+
+      // Similarity: shared tags or shared title keywords
+      const sharedTags = other.tags.filter(t => issueTags.has(t)).length;
+      const otherWords = other.title.toLowerCase().split(/\s+/);
+      const sharedWords = otherWords.filter(w => issueWords.has(w) && w.length > 3).length;
+
+      if (sharedTags > 0 || sharedWords >= 2) {
+        batch.push(other);
+        used.add(other.id);
+      }
+    }
+    batches.push(batch);
+  }
+  return batches;
+}
+
+const batches = groupBySimilarity(issues, batchSize);
+console.log(`Processing ${issues.length} issues in ${batches.length} batch(es) (grouped by similarity)`);
+
+TodoWrite({
+  todos: batches.map((_, i) => ({
+    content: `Plan batch ${i+1}`,
+    status: 'pending',
+    activeForm: `Planning batch ${i+1}`
+  }))
+});
+```
+
+### Phase 2: Unified Explore + Plan (issue-plan-agent) - PARALLEL
+
+```javascript
+Bash(`mkdir -p .workflow/issues/solutions`);
+const pendingSelections = [];  // Collect multi-solution issues for user selection
+
+// Build prompts for all batches
+const agentTasks = batches.map((batch, batchIndex) => {
+  const issueList = batch.map(i => `- ${i.id}: ${i.title}${i.tags.length ? ` [${i.tags.join(', ')}]` : ''}`).join('\n');
+  const batchIds = batch.map(i => i.id);
+
+  const issuePrompt = `
+## Plan Issues
+
+**Issues** (grouped by similarity):
+${issueList}
+
+**Project Root**: ${process.cwd()}
+
+### Project Context (MANDATORY - Read Both Files First)
+1. Read: .workflow/project-tech.json (technology stack, architecture, key components)
+2. Read: .workflow/project-guidelines.json (user-defined constraints and conventions)
+
+**CRITICAL**: All solution tasks MUST comply with constraints in project-guidelines.json
+
+### Steps
+1. Fetch: \`ccw issue status <id> --json\`
+2. Load project context (project-tech.json + project-guidelines.json)
+3. **If source=discovery**: Use discovery_context (file, line, snippet, suggested_fix) as planning hints
+4. Explore (ACE) → Plan solution (respecting guidelines)
+5. Register & bind: \`ccw issue bind <id> --solution <file>\`
+
+### Generate Files
+\`.workflow/issues/solutions/{issue-id}.jsonl\` - Solution with tasks (schema: cat .claude/workflows/cli-templates/schemas/solution-schema.json)
+
+### Binding Rules
+- **Single solution**: Auto-bind via \`ccw issue bind <id> --solution <file>\`
+- **Multiple solutions**: Register only, return for user selection
+
+### Return Summary
+\`\`\`json
+{
+  "bound": [{ "issue_id": "...", "solution_id": "...", "task_count": N }],
+  "pending_selection": [{ "issue_id": "...", "solutions": [{ "id": "...", "description": "...", "task_count": N }] }],
+  "conflicts": [{ "file": "...", "issues": [...] }]
+}
+\`\`\`
+`;
+
+  return { batchIndex, batchIds, issuePrompt, batch };
+});
+
+// Launch agents in parallel (max 10 concurrent)
+const MAX_PARALLEL = 10;
+for (let i = 0; i < agentTasks.length; i += MAX_PARALLEL) {
+  const chunk = agentTasks.slice(i, i + MAX_PARALLEL);
+  const taskIds = [];
+
+  // Launch chunk in parallel
+  for (const { batchIndex, batchIds, issuePrompt, batch } of chunk) {
+    updateTodo(`Plan batch ${batchIndex + 1}`, 'in_progress');
+    const taskId = Task(
+      subagent_type="issue-plan-agent",
+      run_in_background=true,
+      description=`Explore & plan ${batch.length} issues: ${batchIds.join(', ')}`,
+      prompt=issuePrompt
+    );
+    taskIds.push({ taskId, batchIndex });
+  }
+
+  console.log(`Launched ${taskIds.length} agents (batch ${i/MAX_PARALLEL + 1}/${Math.ceil(agentTasks.length/MAX_PARALLEL)})...`);
+
+  // Collect results from this chunk
+  for (const { taskId, batchIndex } of taskIds) {
+    const result = TaskOutput(task_id=taskId, block=true);
+    const summary = JSON.parse(result);
+
+    for (const item of summary.bound || []) {
+      console.log(`✓ ${item.issue_id}: ${item.solution_id} (${item.task_count} tasks)`);
+    }
+    // Collect and notify pending selections
+    for (const pending of summary.pending_selection || []) {
+      console.log(`⏳ ${pending.issue_id}: ${pending.solutions.length} solutions → awaiting selection`);
+      pendingSelections.push(pending);
+    }
+    if (summary.conflicts?.length > 0) {
+      console.log(`⚠ Conflicts: ${summary.conflicts.map(c => c.file).join(', ')}`);
+    }
+    updateTodo(`Plan batch ${batchIndex + 1}`, 'completed');
+  }
+}
+```
+
+### Phase 3: Multi-Solution Selection (MANDATORY when pendingSelections > 0)
+
+```javascript
+// MUST trigger user selection when multiple solutions exist
+if (pendingSelections.length > 0) {
+  console.log(`\n## User Selection Required: ${pendingSelections.length} issue(s) have multiple solutions\n`);
+
+  const answer = AskUserQuestion({
+    questions: pendingSelections.map(({ issue_id, solutions }) => ({
+      question: `Select solution for ${issue_id}:`,
+      header: issue_id,
+      multiSelect: false,
+      options: solutions.map(s => ({
+        label: `${s.id} (${s.task_count} tasks)`,
+        description: s.description
+      }))
+    }))
+  });
+
+  // Bind user-selected solutions
+  for (const { issue_id } of pendingSelections) {
+    const selectedId = extractSelectedSolutionId(answer, issue_id);
+    if (selectedId) {
+      Bash(`ccw issue bind ${issue_id} ${selectedId}`);
+      console.log(`✓ ${issue_id}: ${selectedId} bound`);
+    }
+  }
+}
+```
+
+### Phase 4: Summary
+
+```javascript
+// Count planned issues via CLI
+const plannedIds = Bash(`ccw issue list --status planned --ids`).trim();
+const plannedCount = plannedIds ? plannedIds.split('\n').length : 0;
+
+console.log(`
+## Done: ${issues.length} issues → ${plannedCount} planned
+
+Next: \`/issue:queue\` → \`/issue:execute\`
+`);
+```
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| Issue not found | Auto-create in issues.jsonl |
+| ACE search fails | Agent falls back to ripgrep |
+| No solutions generated | Display error, suggest manual planning |
+| User cancels selection | Skip issue, continue with others |
+| File conflicts | Agent detects and suggests resolution order |
+
+## Related Commands
+
+- `/issue:queue` - Form execution queue from bound solutions
+- `/issue:execute` - Execute queue with codex
+- `ccw issue list` - List all issues
+- `ccw issue status` - View issue and solution details

.claude/commands/issue/queue.md

@@ -0,0 +1,368 @@
+---
+name: queue
+description: Form execution queue from bound solutions using issue-queue-agent (solution-level)
+argument-hint: "[--rebuild] [--issue <id>]"
+allowed-tools: TodoWrite(*), Task(*), Bash(*), Read(*), Write(*)
+---
+
+# Issue Queue Command (/issue:queue)
+
+## Overview
+
+Queue formation command using **issue-queue-agent** that analyzes all bound solutions, resolves **inter-solution** conflicts, and creates an ordered execution queue at **solution level**.
+
+**Design Principle**: Queue items are **solutions**, not individual tasks. Each executor receives a complete solution with all its tasks.
+
+## Output Requirements
+
+**Generate Files:**
+1. `.workflow/issues/queues/{queue-id}.json` - Full queue with solutions, conflicts, groups
+2. `.workflow/issues/queues/index.json` - Update with new queue entry
+
+**Return Summary:**
+```json
+{
+  "queue_id": "QUE-20251227-143000",
+  "total_solutions": N,
+  "total_tasks": N,
+  "execution_groups": [{ "id": "P1", "type": "parallel", "count": N }],
+  "conflicts_resolved": N,
+  "issues_queued": ["ISS-xxx", "ISS-yyy"]
+}
+```
+
+**Completion Criteria:**
+- [ ] Queue JSON generated with valid DAG (no cycles between solutions)
+- [ ] All inter-solution file conflicts resolved with rationale
+- [ ] Semantic priority calculated for each solution
+- [ ] Execution groups assigned (parallel P* / sequential S*)
+- [ ] Issue statuses updated to `queued` via `ccw issue update`
+
+## Core Capabilities
+
+- **Agent-driven**: issue-queue-agent handles all ordering logic
+- **Solution-level granularity**: Queue items are solutions, not tasks
+- Inter-solution dependency DAG (based on file conflicts)
+- File conflict detection between solutions
+- Semantic priority calculation per solution (0.0-1.0)
+- Parallel/Sequential group assignment for solutions
+
+## Storage Structure (Queue History)
+
+```
+.workflow/issues/
+├── issues.jsonl              # All issues (one per line)
+├── queues/                   # Queue history directory
+│   ├── index.json            # Queue index (active + history)
+│   ├── {queue-id}.json       # Individual queue files
+│   └── ...
+└── solutions/
+    ├── {issue-id}.jsonl      # Solutions for issue
+    └── ...
+```
+
+### Queue Index Schema
+
+```json
+{
+  "active_queue_id": "QUE-20251227-143000",
+  "queues": [
+    {
+      "id": "QUE-20251227-143000",
+      "status": "active",
+      "issue_ids": ["ISS-xxx", "ISS-yyy"],
+      "total_solutions": 3,
+      "completed_solutions": 1,
+      "created_at": "2025-12-27T14:30:00Z"
+    }
+  ]
+}
+```
+
+### Queue File Schema (Solution-Level)
+
+```json
+{
+  "id": "QUE-20251227-143000",
+  "status": "active",
+  "solutions": [
+    {
+      "item_id": "S-1",
+      "issue_id": "ISS-20251227-003",
+      "solution_id": "SOL-20251227-003",
+      "status": "pending",
+      "execution_order": 1,
+      "execution_group": "P1",
+      "depends_on": [],
+      "semantic_priority": 0.8,
+      "assigned_executor": "codex",
+      "files_touched": ["src/auth.ts", "src/utils.ts"],
+      "task_count": 3
+    },
+    {
+      "item_id": "S-2",
+      "issue_id": "ISS-20251227-001",
+      "solution_id": "SOL-20251227-001",
+      "status": "pending",
+      "execution_order": 2,
+      "execution_group": "P1",
+      "depends_on": [],
+      "semantic_priority": 0.7,
+      "assigned_executor": "codex",
+      "files_touched": ["src/api.ts"],
+      "task_count": 2
+    },
+    {
+      "item_id": "S-3",
+      "issue_id": "ISS-20251227-002",
+      "solution_id": "SOL-20251227-002",
+      "status": "pending",
+      "execution_order": 3,
+      "execution_group": "S2",
+      "depends_on": ["S-1"],
+      "semantic_priority": 0.5,
+      "assigned_executor": "codex",
+      "files_touched": ["src/auth.ts"],
+      "task_count": 4
+    }
+  ],
+  "conflicts": [
+    {
+      "type": "file_conflict",
+      "file": "src/auth.ts",
+      "solutions": ["S-1", "S-3"],
+      "resolution": "sequential",
+      "resolution_order": ["S-1", "S-3"],
+      "rationale": "S-1 creates auth module, S-3 extends it"
+    }
+  ],
+  "execution_groups": [
+    { "id": "P1", "type": "parallel", "solutions": ["S-1", "S-2"] },
+    { "id": "S2", "type": "sequential", "solutions": ["S-3"] }
+  ]
+}
+```
+
+## Usage
+
+```bash
+/issue:queue [FLAGS]
+
+# Examples
+/issue:queue                      # Form NEW queue from all bound solutions
+/issue:queue --issue GH-123       # Form queue for specific issue only
+/issue:queue --append GH-124      # Append to active queue
+/issue:queue --list               # List all queues (history)
+/issue:queue --switch QUE-xxx     # Switch active queue
+/issue:queue --archive            # Archive completed active queue
+
+# Flags
+--issue <id>          Form queue for specific issue only
+--append <id>         Append issue to active queue (don't create new)
+
+# CLI subcommands (ccw issue queue ...)
+ccw issue queue list                  List all queues with status
+ccw issue queue switch <queue-id>     Switch active queue
+ccw issue queue archive               Archive current queue
+ccw issue queue delete <queue-id>     Delete queue from history
+```
+
+## Execution Process
+
+```
+Phase 1: Solution Loading
+   ├─ Load issues.jsonl
+   ├─ Filter issues with bound_solution_id
+   ├─ Read solutions/{issue-id}.jsonl for each issue
+   ├─ Find bound solution by ID
+   ├─ Collect files_touched from all tasks in solution
+   └─ Build solution objects (NOT individual tasks)
+
+Phase 2-4: Agent-Driven Queue Formation (issue-queue-agent)
+   ├─ Launch issue-queue-agent with all solutions
+   ├─ Agent performs:
+   │   ├─ Detect file overlaps between solutions
+   │   ├─ Build dependency DAG from file conflicts
+   │   ├─ Detect circular dependencies
+   │   ├─ Resolve conflicts using priority rules
+   │   ├─ Calculate semantic priority per solution
+   │   └─ Assign execution groups (parallel/sequential)
+   └─ Output: queue JSON with ordered solutions (S-1, S-2, ...)
+
+Phase 5: Queue Output
+   ├─ Write queue.json with solutions array
+   ├─ Update issue statuses in issues.jsonl
+   └─ Display queue summary
+```
+
+## Implementation
+
+### Phase 1: Solution Loading
+
+**NOTE**: Execute code directly. DO NOT pre-read solution files - Bash cat handles all reading.
+
+```javascript
+// Load issues.jsonl
+const issuesPath = '.workflow/issues/issues.jsonl';
+const allIssues = Bash(`cat "${issuesPath}" 2>/dev/null || echo ''`)
+  .split('\n')
+  .filter(line => line.trim())
+  .map(line => JSON.parse(line));
+
+// Filter issues with bound solutions
+const plannedIssues = allIssues.filter(i =>
+  i.status === 'planned' && i.bound_solution_id
+);
+
+if (plannedIssues.length === 0) {
+  console.log('No issues with bound solutions found.');
+  console.log('Run /issue:plan first to create and bind solutions.');
+  return;
+}
+
+// Load bound solutions (not individual tasks)
+const allSolutions = [];
+for (const issue of plannedIssues) {
+  const solPath = `.workflow/issues/solutions/${issue.id}.jsonl`;
+  const solutions = Bash(`cat "${solPath}" 2>/dev/null || echo ''`)
+    .split('\n')
+    .filter(line => line.trim())
+    .map(line => JSON.parse(line));
+
+  // Find bound solution
+  const boundSol = solutions.find(s => s.id === issue.bound_solution_id);
+
+  if (!boundSol) {
+    console.log(`⚠ Bound solution ${issue.bound_solution_id} not found for ${issue.id}`);
+    continue;
+  }
+
+  // Collect all files touched by this solution
+  const filesTouched = new Set();
+  for (const task of boundSol.tasks || []) {
+    for (const mp of task.modification_points || []) {
+      filesTouched.add(mp.file);
+    }
+  }
+
+  allSolutions.push({
+    issue_id: issue.id,
+    solution_id: issue.bound_solution_id,
+    task_count: boundSol.tasks?.length || 0,
+    files_touched: Array.from(filesTouched),
+    priority: issue.priority || 'medium'
+  });
+}
+
+console.log(`Loaded ${allSolutions.length} solutions from ${plannedIssues.length} issues`);
+```
+
+### Phase 2-4: Agent-Driven Queue Formation
+
+```javascript
+// Generate queue-id ONCE here, pass to agent
+const now = new Date();
+const queueId = `QUE-${now.toISOString().replace(/[-:T]/g, '').slice(0, 14)}`;
+
+// Build minimal prompt - agent orders SOLUTIONS, not tasks
+const agentPrompt = `
+## Order Solutions
+
+**Queue ID**: ${queueId}
+**Solutions**: ${allSolutions.length} from ${plannedIssues.length} issues
+**Project Root**: ${process.cwd()}
+
+### Input (Solution-Level)
+\`\`\`json
+${JSON.stringify(allSolutions, null, 2)}
+\`\`\`
+
+### Steps
+1. Parse solutions: Extract solution IDs, files_touched, task_count, priority
+2. Detect conflicts: Find file overlaps between solutions (files_touched intersection)
+3. Build DAG: Create dependency edges where solutions share files
+4. Detect cycles: Verify no circular dependencies (abort if found)
+5. Resolve conflicts: Apply ordering rules based on action types
+6. Calculate priority: Compute semantic priority (0.0-1.0) per solution
+7. Assign groups: Parallel (P*) for no-conflict, Sequential (S*) for conflicts
+8. Generate queue: Write queue JSON with ordered solutions
+9. Update index: Update queues/index.json with new queue entry
+
+### Rules
+- **Solution Granularity**: Queue items are solutions, NOT individual tasks
+- **DAG Validity**: Output must be valid DAG with no circular dependencies
+- **Conflict Detection**: Two solutions conflict if files_touched intersect
+- **Ordering Priority**:
+  1. Higher issue priority first (critical > high > medium > low)
+  2. Fewer dependencies first (foundation solutions)
+  3. More tasks = higher priority (larger impact)
+- **Parallel Safety**: Solutions in same parallel group must have NO file overlaps
+- **Queue Item ID Format**: \`S-N\` (S-1, S-2, S-3, ...)
+- **Queue ID**: Use the provided Queue ID (passed above), do NOT generate new one
+
+### Generate Files (STRICT - only these 2)
+1. \`.workflow/issues/queues/{Queue ID}.json\` - Use Queue ID from above
+2. \`.workflow/issues/queues/index.json\` - Update existing index
+
+Write ONLY these 2 files, using the provided Queue ID.
+
+### Return Summary
+\`\`\`json
+{
+  "queue_id": "QUE-YYYYMMDD-HHMMSS",
+  "total_solutions": N,
+  "total_tasks": N,
+  "execution_groups": [{ "id": "P1", "type": "parallel", "count": N }],
+  "conflicts_resolved": N,
+  "issues_queued": ["ISS-xxx"]
+}
+\`\`\`
+`;
+
+const result = Task(
+  subagent_type="issue-queue-agent",
+  run_in_background=false,
+  description=`Order ${allSolutions.length} solutions`,
+  prompt=agentPrompt
+);
+
+const summary = JSON.parse(result);
+```
+
+### Phase 5: Summary & Status Update
+
+```javascript
+// Agent already generated queue files, use summary
+console.log(`
+## Queue Formed: ${summary.queue_id}
+
+**Solutions**: ${summary.total_solutions}
+**Tasks**: ${summary.total_tasks}
+**Issues**: ${summary.issues_queued.join(', ')}
+**Groups**: ${summary.execution_groups.map(g => `${g.id}(${g.count})`).join(', ')}
+**Conflicts Resolved**: ${summary.conflicts_resolved}
+
+Next: \`/issue:execute\`
+`);
+
+// Update issue statuses via CLI (use `update` for pure field changes)
+// Note: `queue add` has its own logic; here we only need status update
+for (const issueId of summary.issues_queued) {
+  Bash(`ccw issue update ${issueId} --status queued`);
+}
+```
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| No bound solutions | Display message, suggest /issue:plan |
+| Circular dependency | List cycles, abort queue formation |
+| Unresolved conflicts | Agent resolves using ordering rules |
+| Invalid task reference | Skip and warn |
+
+## Related Commands
+
+- `/issue:plan` - Plan issues and bind solutions
+- `/issue:execute` - Execute queue with codex
+- `ccw issue queue list` - View current queue

.claude/commands/memory/code-map-memory.md

@@ -0,0 +1,687 @@
+---
+name: code-map-memory
+description: 3-phase orchestrator: parse feature keyword → cli-explore-agent analyzes (Deep Scan dual-source) → orchestrator generates Mermaid docs + SKILL package (skips phase 2 if exists)
+argument-hint: "\"feature-keyword\" [--regenerate] [--tool <gemini|qwen>]"
+allowed-tools: SlashCommand(*), TodoWrite(*), Bash(*), Read(*), Write(*), Task(*)
+---
+
+# Code Flow Mapping Generator
+
+## Overview
+
+**Pure Orchestrator with Agent Delegation**: Prepares context paths and delegates code flow analysis to specialized cli-explore-agent. Orchestrator transforms agent's JSON analysis into Mermaid documentation.
+
+**Auto-Continue Workflow**: Runs fully autonomously once triggered. Each phase completes and automatically triggers the next phase.
+
+**Execution Paths**:
+- **Full Path**: All 3 phases (no existing codemap OR `--regenerate` specified)
+- **Skip Path**: Phase 1 → Phase 3 (existing codemap found AND no `--regenerate` flag)
+- **Phase 3 Always Executes**: SKILL index is always generated or updated
+
+**Agent Responsibility** (cli-explore-agent):
+- Deep code flow analysis using dual-source strategy (Bash + Gemini CLI)
+- Returns structured JSON with architecture, functions, data flow, conditionals, patterns
+- NO file writing - analysis only
+
+**Orchestrator Responsibility**:
+- Provides feature keyword and analysis scope to agent
+- Transforms agent's JSON into Mermaid-enriched markdown documentation
+- Writes all files (5 docs + metadata.json + SKILL.md)
+
+## Core Rules
+
+1. **Start Immediately**: First action is TodoWrite initialization, second action is Phase 1 execution
+2. **Feature-Specific SKILL**: Each feature creates independent `.claude/skills/codemap-{feature}/` package
+3. **Specialized Agent**: Phase 2a uses cli-explore-agent for professional code analysis (Deep Scan mode)
+4. **Orchestrator Documentation**: Phase 2b transforms agent JSON into Mermaid markdown files
+5. **Auto-Continue**: After completing each phase, update TodoWrite and immediately execute next phase
+6. **No User Prompts**: Never ask user questions or wait for input between phases
+7. **Track Progress**: Update TodoWrite after EVERY phase completion before starting next phase
+8. **Multi-Level Detail**: Generate 4 levels: architecture → function → data → conditional
+
+---
+
+## 3-Phase Execution
+
+### Phase 1: Parse Feature Keyword & Check Existing
+
+**Goal**: Normalize feature keyword, check existing codemap, prepare for analysis
+
+**Step 1: Parse Feature Keyword**
+```bash
+# Get feature keyword from argument
+FEATURE_KEYWORD="$1"
+
+# Normalize: lowercase, spaces to hyphens
+normalized_feature=$(echo "$FEATURE_KEYWORD" | tr '[:upper:]' '[:lower:]' | tr ' ' '-' | tr '_' '-')
+
+# Example: "User Authentication" → "user-authentication"
+# Example: "支付处理" → "支付处理" (keep non-ASCII)
+```
+
+**Step 2: Set Tool Preference**
+```bash
+# Default to gemini unless --tool specified
+TOOL="${tool_flag:-gemini}"
+```
+
+**Step 3: Check Existing Codemap**
+```bash
+# Define codemap directory
+CODEMAP_DIR=".claude/skills/codemap-${normalized_feature}"
+
+# Check if codemap exists
+bash(test -d "$CODEMAP_DIR" && echo "exists" || echo "not_exists")
+
+# Count existing files
+bash(find "$CODEMAP_DIR" -name "*.md" 2>/dev/null | wc -l || echo 0)
+```
+
+**Step 4: Skip Decision**
+```javascript
+if (existing_files > 0 && !regenerate_flag) {
+  SKIP_GENERATION = true
+  message = "Codemap already exists, skipping Phase 2. Use --regenerate to force regeneration."
+} else if (regenerate_flag) {
+  bash(rm -rf "$CODEMAP_DIR")
+  SKIP_GENERATION = false
+  message = "Regenerating codemap from scratch."
+} else {
+  SKIP_GENERATION = false
+  message = "No existing codemap found, generating new code flow analysis."
+}
+```
+
+**Output Variables**:
+- `FEATURE_KEYWORD`: Original feature keyword
+- `normalized_feature`: Normalized feature name for directory
+- `CODEMAP_DIR`: `.claude/skills/codemap-{feature}`
+- `TOOL`: CLI tool to use (gemini or qwen)
+- `SKIP_GENERATION`: Boolean - whether to skip Phase 2
+
+**TodoWrite**:
+- If skipping: Mark phase 1 completed, phase 2 completed, phase 3 in_progress
+- If not skipping: Mark phase 1 completed, phase 2 in_progress
+
+---
+
+### Phase 2: Code Flow Analysis & Documentation Generation
+
+**Skip Condition**: Skipped if `SKIP_GENERATION = true`
+
+**Goal**: Use cli-explore-agent for professional code analysis, then orchestrator generates Mermaid documentation
+
+**Architecture**: Phase 2a (Agent Analysis) → Phase 2b (Orchestrator Documentation)
+
+---
+
+#### Phase 2a: cli-explore-agent Analysis
+
+**Purpose**: Leverage specialized cli-explore-agent for deep code flow analysis
+
+**Agent Task Specification**:
+
+```
+Task(
+  subagent_type: "cli-explore-agent",
+  description: "Analyze code flow: {FEATURE_KEYWORD}",
+  prompt: "
+Perform Deep Scan analysis for feature: {FEATURE_KEYWORD}
+
+**Analysis Mode**: deep-scan (Dual-source: Bash structural scan + Gemini semantic analysis)
+
+**Analysis Objectives**:
+1. **Module Architecture**: Identify high-level module organization, interactions, and entry points
+2. **Function Call Chains**: Trace execution paths, call sequences, and parameter flows
+3. **Data Transformations**: Map data structure changes and transformation stages
+4. **Conditional Paths**: Document decision trees, branches, and error handling strategies
+5. **Design Patterns**: Discover architectural patterns and extract design intent
+
+**Scope**:
+- Feature: {FEATURE_KEYWORD}
+- CLI Tool: {TOOL} (gemini-2.5-pro or qwen coder-model)
+- File Discovery: MCP Code Index (preferred) + rg fallback
+- Target: 5-15 most relevant files
+
+**MANDATORY FIRST STEP**:
+Read: ~/.claude/workflows/cli-templates/schemas/codemap-json-schema.json
+
+**Output**: Return JSON following schema exactly. NO FILE WRITING - return JSON analysis only.
+
+**Critical Requirements**:
+- Use Deep Scan mode: Bash (Phase 1 - precise locations) + Gemini CLI (Phase 2 - semantic understanding) + Synthesis (Phase 3 - merge with attribution)
+- Focus exclusively on {FEATURE_KEYWORD} feature flow
+- Include file:line references for ALL findings
+- Extract design intent from code structure and comments
+- NO FILE WRITING - return JSON analysis only
+- Handle tool failures gracefully (Gemini → Qwen fallback, MCP → rg fallback)
+  "
+)
+```
+
+**Agent Output**: JSON analysis result with architecture, functions, data flow, conditionals, and patterns
+
+---
+
+#### Phase 2b: Orchestrator Documentation Generation
+
+**Purpose**: Transform cli-explore-agent JSON into Mermaid-enriched documentation
+
+**Input**: Agent's JSON analysis result
+
+**Process**:
+
+1. **Parse Agent Analysis**:
+   ```javascript
+   const analysis = JSON.parse(agentResult)
+   const { feature, files_analyzed, architecture, function_calls, data_flow, conditional_logic, design_patterns } = analysis
+   ```
+
+2. **Generate Mermaid Diagrams from Structured Data**:
+
+   **a) architecture-flow.md** (~3K tokens):
+   ```javascript
+   // Convert architecture.modules + architecture.interactions → Mermaid graph TD
+   const architectureMermaid = `
+   graph TD
+   ${architecture.modules.map(m => `    ${m.name}[${m.name}]`).join('\n')}
+   ${architecture.interactions.map(i => `    ${i.from} -->|${i.type}| ${i.to}`).join('\n')}
+   `
+
+   Write({
+     file_path: `${CODEMAP_DIR}/architecture-flow.md`,
+     content: `---
+feature: ${feature}
+level: architecture
+detail: high-level module interactions
+---
+# Architecture Flow: ${feature}
+
+## Overview
+${architecture.overview}
+
+## Module Architecture
+${architecture.modules.map(m => `### ${m.name}\n- **File**: ${m.file}\n- **Role**: ${m.responsibility}\n- **Dependencies**: ${m.dependencies.join(', ')}`).join('\n\n')}
+
+## Flow Diagram
+\`\`\`mermaid
+${architectureMermaid}
+\`\`\`
+
+## Key Interactions
+${architecture.interactions.map(i => `- **${i.from} → ${i.to}**: ${i.description}`).join('\n')}
+
+## Entry Points
+${architecture.entry_points.map(e => `- **${e.function}** (${e.file}): ${e.description}`).join('\n')}
+`
+   })
+   ```
+
+   **b) function-calls.md** (~5K tokens):
+   ```javascript
+   // Convert function_calls.sequences → Mermaid sequenceDiagram
+   const sequenceMermaid = `
+   sequenceDiagram
+   ${function_calls.sequences.map(s => `    ${s.from}->>${s.to}: ${s.method}`).join('\n')}
+   `
+
+   Write({
+     file_path: `${CODEMAP_DIR}/function-calls.md`,
+     content: `---
+feature: ${feature}
+level: function
+detail: function-level call sequences
+---
+# Function Call Chains: ${feature}
+
+## Call Sequence Diagram
+\`\`\`mermaid
+${sequenceMermaid}
+\`\`\`
+
+## Detailed Call Chains
+${function_calls.call_chains.map(chain => `
+### Chain ${chain.chain_id}: ${chain.description}
+${chain.sequence.map(fn => `- **${fn.function}** (${fn.file})\n  - Calls: ${fn.calls.join(', ')}`).join('\n')}
+`).join('\n')}
+
+## Parameters & Returns
+${function_calls.sequences.map(s => `- **${s.method}** → Returns: ${s.returns || 'void'}`).join('\n')}
+`
+   })
+   ```
+
+   **c) data-flow.md** (~4K tokens):
+   ```javascript
+   // Convert data_flow.transformations → Mermaid flowchart LR
+   const dataFlowMermaid = `
+   flowchart LR
+   ${data_flow.transformations.map((t, i) => `    Stage${i}[${t.from}] -->|${t.transformer}| Stage${i+1}[${t.to}]`).join('\n')}
+   `
+
+   Write({
+     file_path: `${CODEMAP_DIR}/data-flow.md`,
+     content: `---
+feature: ${feature}
+level: data
+detail: data structure transformations
+---
+# Data Flow: ${feature}
+
+## Data Transformation Diagram
+\`\`\`mermaid
+${dataFlowMermaid}
+\`\`\`
+
+## Data Structures
+${data_flow.structures.map(s => `### ${s.name} (${s.stage})\n\`\`\`json\n${JSON.stringify(s.shape, null, 2)}\n\`\`\``).join('\n\n')}
+
+## Transformations
+${data_flow.transformations.map(t => `- **${t.from} → ${t.to}** via \`${t.transformer}\` (${t.file})`).join('\n')}
+`
+   })
+   ```
+
+   **d) conditional-paths.md** (~4K tokens):
+   ```javascript
+   // Convert conditional_logic.branches → Mermaid flowchart TD
+   const conditionalMermaid = `
+   flowchart TD
+       Start[Entry Point]
+   ${conditional_logic.branches.map((b, i) => `
+       Start --> Check${i}{${b.condition}}
+       Check${i} -->|Yes| Path${i}A[${b.true_path}]
+       Check${i} -->|No| Path${i}B[${b.false_path}]
+   `).join('\n')}
+   `
+
+   Write({
+     file_path: `${CODEMAP_DIR}/conditional-paths.md`,
+     content: `---
+feature: ${feature}
+level: conditional
+detail: decision trees and error paths
+---
+# Conditional Paths: ${feature}
+
+## Decision Tree
+\`\`\`mermaid
+${conditionalMermaid}
+\`\`\`
+
+## Branch Conditions
+${conditional_logic.branches.map(b => `- **${b.condition}** (${b.file})\n  - True: ${b.true_path}\n  - False: ${b.false_path}`).join('\n')}
+
+## Error Handling
+${conditional_logic.error_handling.map(e => `- **${e.error_type}**: Handler \`${e.handler}\` (${e.file}) - Recovery: ${e.recovery}`).join('\n')}
+`
+   })
+   ```
+
+   **e) complete-flow.md** (~8K tokens):
+   ```javascript
+   // Integrate all Mermaid diagrams
+   Write({
+     file_path: `${CODEMAP_DIR}/complete-flow.md`,
+     content: `---
+feature: ${feature}
+level: complete
+detail: integrated multi-level view
+---
+# Complete Flow: ${feature}
+
+## Integrated Flow Diagram
+\`\`\`mermaid
+graph TB
+    subgraph Architecture
+    ${architecture.modules.map(m => `    ${m.name}[${m.name}]`).join('\n')}
+    end
+
+    subgraph "Function Calls"
+    ${function_calls.call_chains[0]?.sequence.map(fn => `    ${fn.function}`).join('\n') || ''}
+    end
+
+    subgraph "Data Flow"
+    ${data_flow.structures.map(s => `    ${s.name}[${s.name}]`).join('\n')}
+    end
+\`\`\`
+
+## Complete Trace
+[Comprehensive end-to-end documentation combining all analysis layers]
+
+## Design Patterns Identified
+${design_patterns.map(p => `- **${p.pattern}** in ${p.location}: ${p.description}`).join('\n')}
+
+## Recommendations
+${analysis.recommendations.map(r => `- ${r}`).join('\n')}
+
+## Cross-References
+- [Architecture Flow](./architecture-flow.md) - High-level module structure
+- [Function Calls](./function-calls.md) - Detailed call chains
+- [Data Flow](./data-flow.md) - Data transformation stages
+- [Conditional Paths](./conditional-paths.md) - Decision trees and error handling
+`
+   })
+   ```
+
+3. **Write metadata.json**:
+   ```javascript
+   Write({
+     file_path: `${CODEMAP_DIR}/metadata.json`,
+     content: JSON.stringify({
+       feature: feature,
+       normalized_name: normalized_feature,
+       generated_at: new Date().toISOString(),
+       tool_used: analysis.analysis_metadata.tool_used,
+       files_analyzed: files_analyzed.map(f => f.file),
+       analysis_summary: {
+         total_files: files_analyzed.length,
+         modules_traced: architecture.modules.length,
+         functions_traced: function_calls.call_chains.reduce((sum, c) => sum + c.sequence.length, 0),
+         patterns_discovered: design_patterns.length
+       }
+     }, null, 2)
+   })
+   ```
+
+4. **Report Phase 2 Completion**:
+   ```
+   Phase 2 Complete: Code flow analysis and documentation generated
+
+   - Agent Analysis: cli-explore-agent with {TOOL}
+   - Files Analyzed: {count}
+   - Documentation Generated: 5 markdown files + metadata.json
+   - Location: {CODEMAP_DIR}
+   ```
+
+**Completion Criteria**:
+- cli-explore-agent task completed successfully with JSON result
+- 5 documentation files written with valid Mermaid diagrams
+- metadata.json written with analysis summary
+- All files properly formatted and cross-referenced
+
+**TodoWrite**: Mark phase 2 completed, phase 3 in_progress
+
+---
+
+### Phase 3: Generate SKILL.md Index
+
+**Note**: This phase **ALWAYS executes** - generates or updates the SKILL index.
+
+**Goal**: Read generated flow documentation and create SKILL.md index with progressive loading
+
+**Steps**:
+
+1. **Verify Generated Files**:
+   ```bash
+   bash(find "{CODEMAP_DIR}" -name "*.md" -type f | sort)
+   ```
+
+2. **Read metadata.json**:
+   ```javascript
+   Read({CODEMAP_DIR}/metadata.json)
+   // Extract: feature, normalized_name, files_analyzed, analysis_summary
+   ```
+
+3. **Read File Headers** (optional, first 30 lines):
+   ```javascript
+   Read({CODEMAP_DIR}/architecture-flow.md, limit: 30)
+   Read({CODEMAP_DIR}/function-calls.md, limit: 30)
+   // Extract overview and diagram counts
+   ```
+
+4. **Generate SKILL.md Index**:
+
+   Template structure:
+   ```yaml
+   ---
+   name: codemap-{normalized_feature}
+   description: Code flow mapping for {FEATURE_KEYWORD} feature (located at {project_path}). Load this SKILL when analyzing, tracing, or understanding {FEATURE_KEYWORD} execution flow, especially when no relevant context exists in memory.
+   version: 1.0.0
+   generated_at: {ISO_TIMESTAMP}
+   ---
+   # Code Flow Map: {FEATURE_KEYWORD}
+
+   ## Feature: `{FEATURE_KEYWORD}`
+
+   **Analysis Date**: {DATE}
+   **Tool Used**: {TOOL}
+   **Files Analyzed**: {COUNT}
+
+   ## Progressive Loading
+
+   ### Level 0: Quick Overview (~2K tokens)
+   - [Architecture Flow](./architecture-flow.md) - High-level module interactions
+
+   ### Level 1: Core Flows (~10K tokens)
+   - [Architecture Flow](./architecture-flow.md) - Module architecture
+   - [Function Calls](./function-calls.md) - Function call chains
+
+   ### Level 2: Complete Analysis (~20K tokens)
+   - [Architecture Flow](./architecture-flow.md)
+   - [Function Calls](./function-calls.md)
+   - [Data Flow](./data-flow.md) - Data transformations
+
+   ### Level 3: Deep Dive (~30K tokens)
+   - [Architecture Flow](./architecture-flow.md)
+   - [Function Calls](./function-calls.md)
+   - [Data Flow](./data-flow.md)
+   - [Conditional Paths](./conditional-paths.md) - Branches and error handling
+   - [Complete Flow](./complete-flow.md) - Integrated comprehensive view
+
+   ## Usage
+
+   Load this SKILL package when:
+   - Analyzing {FEATURE_KEYWORD} implementation
+   - Tracing execution flow for debugging
+   - Understanding code dependencies
+   - Planning refactoring or enhancements
+
+   ## Analysis Summary
+
+   - **Modules Traced**: {modules_traced}
+   - **Functions Traced**: {functions_traced}
+   - **Files Analyzed**: {total_files}
+
+   ## Mermaid Diagrams Included
+
+   - Architecture flow diagram (graph TD)
+   - Function call sequence diagram (sequenceDiagram)
+   - Data transformation flowchart (flowchart LR)
+   - Conditional decision tree (flowchart TD)
+   - Complete integrated diagram (graph TB)
+   ```
+
+5. **Write SKILL.md**:
+   ```javascript
+   Write({
+     file_path: `{CODEMAP_DIR}/SKILL.md`,
+     content: generatedIndexMarkdown
+   })
+   ```
+
+**Completion Criteria**:
+- SKILL.md index written
+- All documentation files verified
+- Progressive loading levels (0-3) properly structured
+- Mermaid diagram references included
+
+**TodoWrite**: Mark phase 3 completed
+
+**Final Report**:
+```
+Code Flow Mapping Complete
+
+Feature: {FEATURE_KEYWORD}
+Location: .claude/skills/codemap-{normalized_feature}/
+
+Files Generated:
+- SKILL.md (index)
+- architecture-flow.md (with Mermaid diagram)
+- function-calls.md (with Mermaid sequence diagram)
+- data-flow.md (with Mermaid flowchart)
+- conditional-paths.md (with Mermaid decision tree)
+- complete-flow.md (with integrated Mermaid diagram)
+- metadata.json
+
+Analysis:
+- Files analyzed: {count}
+- Modules traced: {count}
+- Functions traced: {count}
+
+Usage: Skill(command: "codemap-{normalized_feature}")
+```
+
+---
+
+## Implementation Details
+
+### TodoWrite Patterns
+
+**Initialization** (Before Phase 1):
+```javascript
+TodoWrite({todos: [
+  {"content": "Parse feature keyword and check existing", "status": "in_progress", "activeForm": "Parsing feature keyword"},
+  {"content": "Agent analyzes code flow and generates files", "status": "pending", "activeForm": "Analyzing code flow"},
+  {"content": "Generate SKILL.md index", "status": "pending", "activeForm": "Generating SKILL index"}
+]})
+```
+
+**Full Path** (SKIP_GENERATION = false):
+```javascript
+// After Phase 1
+TodoWrite({todos: [
+  {"content": "Parse feature keyword and check existing", "status": "completed", ...},
+  {"content": "Agent analyzes code flow and generates files", "status": "in_progress", ...},
+  {"content": "Generate SKILL.md index", "status": "pending", ...}
+]})
+
+// After Phase 2
+TodoWrite({todos: [
+  {"content": "Parse feature keyword and check existing", "status": "completed", ...},
+  {"content": "Agent analyzes code flow and generates files", "status": "completed", ...},
+  {"content": "Generate SKILL.md index", "status": "in_progress", ...}
+]})
+
+// After Phase 3
+TodoWrite({todos: [
+  {"content": "Parse feature keyword and check existing", "status": "completed", ...},
+  {"content": "Agent analyzes code flow and generates files", "status": "completed", ...},
+  {"content": "Generate SKILL.md index", "status": "completed", ...}
+]})
+```
+
+**Skip Path** (SKIP_GENERATION = true):
+```javascript
+// After Phase 1 (skip Phase 2)
+TodoWrite({todos: [
+  {"content": "Parse feature keyword and check existing", "status": "completed", ...},
+  {"content": "Agent analyzes code flow and generates files", "status": "completed", ...},  // Skipped
+  {"content": "Generate SKILL.md index", "status": "in_progress", ...}
+]})
+```
+
+### Execution Flow
+
+**Full Path**:
+```
+User → TodoWrite Init → Phase 1 (parse) → Phase 2 (agent analyzes) → Phase 3 (write index) → Report
+```
+
+**Skip Path**:
+```
+User → TodoWrite Init → Phase 1 (detect existing) → Phase 3 (update index) → Report
+```
+
+### Error Handling
+
+**Phase 1 Errors**:
+- Empty feature keyword: Report error, ask user to provide feature description
+- Invalid characters: Normalize and continue
+
+**Phase 2 Errors (Agent)**:
+- Agent task fails: Retry once, report if fails again
+- No files discovered: Warn user, ask for more specific feature keyword
+- CLI failures: Agent handles internally with retries
+- Invalid Mermaid syntax: Agent validates before writing
+
+**Phase 3 Errors**:
+- Write failures: Report which files failed
+- Missing files: Note in SKILL.md, suggest regeneration
+
+---
+
+## Parameters
+
+```bash
+/memory:code-map-memory "feature-keyword" [--regenerate] [--tool <gemini|qwen>]
+```
+
+**Arguments**:
+- **"feature-keyword"**: Feature or flow to analyze (required)
+  - Examples: `"user authentication"`, `"payment processing"`, `"数据导入流程"`
+  - Can be English, Chinese, or mixed
+  - Spaces and underscores normalized to hyphens
+- **--regenerate**: Force regenerate existing codemap (deletes and recreates)
+- **--tool**: CLI tool for analysis (default: gemini)
+  - `gemini`: Comprehensive flow analysis with gemini-2.5-pro
+  - `qwen`: Alternative with coder-model
+
+---
+
+## Examples
+
+**Generated File Structure** (for all examples):
+```
+.claude/skills/codemap-{feature}/
+├── SKILL.md                    # Index (Phase 3)
+├── architecture-flow.md        # Agent (Phase 2) - High-level flow
+├── function-calls.md           # Agent (Phase 2) - Function chains
+├── data-flow.md                # Agent (Phase 2) - Data transformations
+├── conditional-paths.md        # Agent (Phase 2) - Branches & errors
+├── complete-flow.md            # Agent (Phase 2) - Integrated view
+└── metadata.json               # Agent (Phase 2)
+```
+
+### Example 1: User Authentication Flow
+
+```bash
+/memory:code-map-memory "user authentication"
+```
+
+**Workflow**:
+1. Phase 1: Normalizes to "user-authentication", checks existing codemap
+2. Phase 2: Agent discovers auth-related files, executes CLI analysis, generates 5 flow docs with Mermaid
+3. Phase 3: Generates SKILL.md index with progressive loading
+
+**Output**: `.claude/skills/codemap-user-authentication/` with 6 files + metadata
+
+
+### Example 3: Regenerate with Qwen
+
+```bash
+/memory:code-map-memory "payment processing" --regenerate --tool qwen
+```
+
+**Workflow**:
+1. Phase 1: Deletes existing codemap due to --regenerate
+2. Phase 2: Agent uses qwen with coder-model for fresh analysis
+3. Phase 3: Generates updated SKILL.md
+
+---
+
+
+## Architecture
+
+```
+code-map-memory (orchestrator)
+  ├─ Phase 1: Parse & Check (bash commands, skip decision)
+  ├─ Phase 2: Code Analysis & Documentation (skippable)
+  │   ├─ Phase 2a: cli-explore-agent Analysis
+  │   │   └─ Deep Scan: Bash structural + Gemini semantic → JSON
+  │   └─ Phase 2b: Orchestrator Documentation
+  │       └─ Transform JSON → 5 Mermaid markdown files + metadata.json
+  └─ Phase 3: Write SKILL.md (index generation, always runs)
+
+Output: .claude/skills/codemap-{feature}/
+```

.claude/commands/memory/compact.md

@@ -0,0 +1,383 @@
+---
+name: compact
+description: Compact current session memory into structured text for session recovery, extracting objective/plan/files/decisions/constraints/state, and save via MCP core_memory tool
+argument-hint: "[optional: session description]"
+allowed-tools: mcp__ccw-tools__core_memory(*), Read(*)
+examples:
+  - /memory:compact
+  - /memory:compact "completed core-memory module"
+---
+
+# Memory Compact Command (/memory:compact)
+
+## 1. Overview
+
+The `memory:compact` command **compresses current session working memory** into structured text optimized for **session recovery**, extracts critical information, and saves it to persistent storage via MCP `core_memory` tool.
+
+**Core Philosophy**:
+- **Session Recovery First**: Capture everything needed to resume work seamlessly
+- **Minimize Re-exploration**: Include file paths, decisions, and state to avoid redundant analysis
+- **Preserve Train of Thought**: Keep notes and hypotheses for complex debugging
+- **Actionable State**: Record last action result and known issues
+
+## 2. Parameters
+
+- `"session description"` (Optional): Session description to supplement objective
+  - Example: "completed core-memory module"
+  - Example: "debugging JWT refresh - suspected memory leak"
+
+## 3. Structured Output Format
+
+```markdown
+## Session ID
+[WFS-ID if workflow session active, otherwise (none)]
+
+## Project Root
+[Absolute path to project root, e.g., D:\Claude_dms3]
+
+## Objective
+[High-level goal - the "North Star" of this session]
+
+## Execution Plan
+[CRITICAL: Embed the LATEST plan in its COMPLETE and DETAILED form]
+
+### Source: [workflow | todo | user-stated | inferred]
+
+<details>
+<summary>Full Execution Plan (Click to expand)</summary>
+
+[PRESERVE COMPLETE PLAN VERBATIM - DO NOT SUMMARIZE]
+- ALL phases, tasks, subtasks
+- ALL file paths (absolute)
+- ALL dependencies and prerequisites
+- ALL acceptance criteria
+- ALL status markers ([x] done, [ ] pending)
+- ALL notes and context
+
+Example:
+## Phase 1: Setup
+- [x] Initialize project structure
+  - Created D:\Claude_dms3\src\core\index.ts
+  - Added dependencies: lodash, zod
+- [ ] Configure TypeScript
+  - Update tsconfig.json for strict mode
+
+## Phase 2: Implementation
+- [ ] Implement core API
+  - Target: D:\Claude_dms3\src\api\handler.ts
+  - Dependencies: Phase 1 complete
+  - Acceptance: All tests pass
+
+</details>
+
+## Working Files (Modified)
+[Absolute paths to actively modified files]
+- D:\Claude_dms3\src\file1.ts (role: main implementation)
+- D:\Claude_dms3\tests\file1.test.ts (role: unit tests)
+
+## Reference Files (Read-Only)
+[Absolute paths to context files - NOT modified but essential for understanding]
+- D:\Claude_dms3\.claude\CLAUDE.md (role: project instructions)
+- D:\Claude_dms3\src\types\index.ts (role: type definitions)
+- D:\Claude_dms3\package.json (role: dependencies)
+
+## Last Action
+[Last significant action and its result/status]
+
+## Decisions
+- [Decision]: [Reasoning]
+- [Decision]: [Reasoning]
+
+## Constraints
+- [User-specified limitation or preference]
+
+## Dependencies
+- [Added/changed packages or environment requirements]
+
+## Known Issues
+- [Deferred bug or edge case]
+
+## Changes Made
+- [Completed modification]
+
+## Pending
+- [Next step] or (none)
+
+## Notes
+[Unstructured thoughts, hypotheses, debugging trails]
+```
+
+## 4. Field Definitions
+
+| Field | Purpose | Recovery Value |
+|-------|---------|----------------|
+| **Session ID** | Workflow session identifier (WFS-*) | Links memory to specific stateful task execution |
+| **Project Root** | Absolute path to project directory | Enables correct path resolution in new sessions |
+| **Objective** | Ultimate goal of the session | Prevents losing track of broader feature |
+| **Execution Plan** | Complete plan from any source (verbatim) | Preserves full planning context, avoids re-planning |
+| **Working Files** | Actively modified files (absolute paths) | Immediately identifies where work was happening |
+| **Reference Files** | Read-only context files (absolute paths) | Eliminates re-exploration for critical context |
+| **Last Action** | Final tool output/status | Immediate state awareness (success/failure) |
+| **Decisions** | Architectural choices + reasoning | Prevents re-litigating settled decisions |
+| **Constraints** | User-imposed limitations | Maintains personalized coding style |
+| **Dependencies** | Package/environment changes | Prevents missing dependency errors |
+| **Known Issues** | Deferred bugs/edge cases | Ensures issues aren't forgotten |
+| **Changes Made** | Completed modifications | Clear record of what was done |
+| **Pending** | Next steps | Immediate action items |
+| **Notes** | Hypotheses, debugging trails | Preserves "train of thought" |
+
+## 5. Execution Flow
+
+### Step 1: Analyze Current Session
+
+Extract the following from conversation history:
+
+```javascript
+const sessionAnalysis = {
+  sessionId: "",       // WFS-* if workflow session active, null otherwise
+  projectRoot: "",     // Absolute path: D:\Claude_dms3
+  objective: "",       // High-level goal (1-2 sentences)
+  executionPlan: {
+    source: "workflow" | "todo" | "user-stated" | "inferred",
+    content: ""        // Full plan content - ALWAYS preserve COMPLETE and DETAILED form
+  },
+  workingFiles: [],    // {absolutePath, role} - modified files
+  referenceFiles: [],  // {absolutePath, role} - read-only context files
+  lastAction: "",      // Last significant action + result
+  decisions: [],       // {decision, reasoning}
+  constraints: [],     // User-specified limitations
+  dependencies: [],    // Added/changed packages
+  knownIssues: [],     // Deferred bugs
+  changesMade: [],     // Completed modifications
+  pending: [],         // Next steps
+  notes: ""            // Unstructured thoughts
+}
+```
+
+### Step 2: Generate Structured Text
+
+```javascript
+// Helper: Generate execution plan section
+const generateExecutionPlan = (plan) => {
+  const sourceLabels = {
+    'workflow': 'workflow (IMPL_PLAN.md)',
+    'todo': 'todo (TodoWrite)',
+    'user-stated': 'user-stated',
+    'inferred': 'inferred'
+  };
+
+  // CRITICAL: Preserve complete plan content verbatim - DO NOT summarize
+  return `### Source: ${sourceLabels[plan.source] || plan.source}
+
+<details>
+<summary>Full Execution Plan (Click to expand)</summary>
+
+${plan.content}
+
+</details>`;
+};
+
+const structuredText = `## Session ID
+${sessionAnalysis.sessionId || '(none)'}
+
+## Project Root
+${sessionAnalysis.projectRoot}
+
+## Objective
+${sessionAnalysis.objective}
+
+## Execution Plan
+${generateExecutionPlan(sessionAnalysis.executionPlan)}
+
+## Working Files (Modified)
+${sessionAnalysis.workingFiles.map(f => `- ${f.absolutePath} (role: ${f.role})`).join('\n') || '(none)'}
+
+## Reference Files (Read-Only)
+${sessionAnalysis.referenceFiles.map(f => `- ${f.absolutePath} (role: ${f.role})`).join('\n') || '(none)'}
+
+## Last Action
+${sessionAnalysis.lastAction}
+
+## Decisions
+${sessionAnalysis.decisions.map(d => `- ${d.decision}: ${d.reasoning}`).join('\n') || '(none)'}
+
+## Constraints
+${sessionAnalysis.constraints.map(c => `- ${c}`).join('\n') || '(none)'}
+
+## Dependencies
+${sessionAnalysis.dependencies.map(d => `- ${d}`).join('\n') || '(none)'}
+
+## Known Issues
+${sessionAnalysis.knownIssues.map(i => `- ${i}`).join('\n') || '(none)'}
+
+## Changes Made
+${sessionAnalysis.changesMade.map(c => `- ${c}`).join('\n') || '(none)'}
+
+## Pending
+${sessionAnalysis.pending.length > 0
+  ? sessionAnalysis.pending.map(p => `- ${p}`).join('\n')
+  : '(none)'}
+
+## Notes
+${sessionAnalysis.notes || '(none)'}`
+```
+
+### Step 3: Import to Core Memory via MCP
+
+Use the MCP `core_memory` tool to save the structured text:
+
+```javascript
+mcp__ccw-tools__core_memory({
+  operation: "import",
+  text: structuredText
+})
+```
+
+Or via CLI (pipe structured text to import):
+
+```bash
+# Write structured text to temp file, then import
+echo "$structuredText" | ccw core-memory import
+
+# Or from a file
+ccw core-memory import --file /path/to/session-memory.md
+```
+
+**Response Format**:
+```json
+{
+  "operation": "import",
+  "id": "CMEM-YYYYMMDD-HHMMSS",
+  "message": "Created memory: CMEM-YYYYMMDD-HHMMSS"
+}
+```
+
+### Step 4: Report Recovery ID
+
+After successful import, **clearly display the Recovery ID** to the user:
+
+```
+╔════════════════════════════════════════════════════════════════════════════╗
+║  ✓ Session Memory Saved                                                    ║
+║                                                                            ║
+║  Recovery ID: CMEM-YYYYMMDD-HHMMSS                                         ║
+║                                                                            ║
+║  To restore: "Please import memory <ID>"                                   ║
+║  (MCP: core_memory export | CLI: ccw core-memory export --id <ID>)         ║
+╚════════════════════════════════════════════════════════════════════════════╝
+```
+
+## 6. Quality Checklist
+
+Before generating:
+- [ ] Session ID captured if workflow session active (WFS-*)
+- [ ] Project Root is absolute path (e.g., D:\Claude_dms3)
+- [ ] Objective clearly states the "North Star" goal
+- [ ] Execution Plan: COMPLETE plan preserved VERBATIM (no summarization)
+- [ ] Plan Source: Clearly identified (workflow | todo | user-stated | inferred)
+- [ ] Plan Details: ALL phases, tasks, file paths, dependencies, status markers included
+- [ ] All file paths are ABSOLUTE (not relative)
+- [ ] Working Files: 3-8 modified files with roles
+- [ ] Reference Files: Key context files (CLAUDE.md, types, configs)
+- [ ] Last Action captures final state (success/failure)
+- [ ] Decisions include reasoning, not just choices
+- [ ] Known Issues separates deferred from forgotten bugs
+- [ ] Notes preserve debugging hypotheses if any
+
+## 7. Path Resolution Rules
+
+### Project Root Detection
+1. Check current working directory from environment
+2. Look for project markers: `.git/`, `package.json`, `.claude/`
+3. Use the topmost directory containing these markers
+
+### Absolute Path Conversion
+```javascript
+// Convert relative to absolute
+const toAbsolutePath = (relativePath, projectRoot) => {
+  if (path.isAbsolute(relativePath)) return relativePath;
+  return path.join(projectRoot, relativePath);
+};
+
+// Example: "src/api/auth.ts" → "D:\Claude_dms3\src\api\auth.ts"
+```
+
+### Reference File Categories
+| Category | Examples | Priority |
+|----------|----------|----------|
+| Project Config | `.claude/CLAUDE.md`, `package.json`, `tsconfig.json` | High |
+| Type Definitions | `src/types/*.ts`, `*.d.ts` | High |
+| Related Modules | Parent/sibling modules with shared interfaces | Medium |
+| Test Files | Corresponding test files for modified code | Medium |
+| Documentation | `README.md`, `ARCHITECTURE.md` | Low |
+
+## 8. Plan Detection (Priority Order)
+
+### Priority 1: Workflow Session (IMPL_PLAN.md)
+```javascript
+// Check for active workflow session
+const manifest = await mcp__ccw-tools__session_manager({
+  operation: "list",
+  location: "active"
+});
+
+if (manifest.sessions?.length > 0) {
+  const session = manifest.sessions[0];
+  const plan = await mcp__ccw-tools__session_manager({
+    operation: "read",
+    session_id: session.id,
+    content_type: "plan"
+  });
+  sessionAnalysis.sessionId = session.id;
+  sessionAnalysis.executionPlan.source = "workflow";
+  sessionAnalysis.executionPlan.content = plan.content;
+}
+```
+
+### Priority 2: TodoWrite (Current Session Todos)
+```javascript
+// Extract from conversation - look for TodoWrite tool calls
+// Preserve COMPLETE todo list with all details
+const todos = extractTodosFromConversation();
+if (todos.length > 0) {
+  sessionAnalysis.executionPlan.source = "todo";
+  // Format todos with full context - preserve status markers
+  sessionAnalysis.executionPlan.content = todos.map(t =>
+    `- [${t.status === 'completed' ? 'x' : t.status === 'in_progress' ? '>' : ' '}] ${t.content}`
+  ).join('\n');
+}
+```
+
+### Priority 3: User-Stated Plan
+```javascript
+// Look for explicit plan statements in user messages:
+// - "Here's my plan: 1. ... 2. ... 3. ..."
+// - "I want to: first..., then..., finally..."
+// - Numbered or bulleted lists describing steps
+const userPlan = extractUserStatedPlan();
+if (userPlan) {
+  sessionAnalysis.executionPlan.source = "user-stated";
+  sessionAnalysis.executionPlan.content = userPlan;
+}
+```
+
+### Priority 4: Inferred Plan
+```javascript
+// If no explicit plan, infer from:
+// - Task description and breakdown discussion
+// - Sequence of actions taken
+// - Outstanding work mentioned
+const inferredPlan = inferPlanFromDiscussion();
+if (inferredPlan) {
+  sessionAnalysis.executionPlan.source = "inferred";
+  sessionAnalysis.executionPlan.content = inferredPlan;
+}
+```
+
+## 9. Notes
+
+- **Timing**: Execute at task completion or before context switch
+- **Frequency**: Once per independent task or milestone
+- **Recovery**: New session can immediately continue with full context
+- **Knowledge Graph**: Entity relationships auto-extracted for visualization
+- **Absolute Paths**: Critical for cross-session recovery on different machines

.claude/commands/memory/docs-full-cli.md

@@ -0,0 +1,471 @@
+---
+name: docs-full-cli
+description: Generate full project documentation using CLI execution (Layer 3→1) with batched agents (4 modules/agent) and gemini→qwen→codex fallback, <20 modules uses direct parallel
+argument-hint: "[path] [--tool <gemini|qwen|codex>]"
+---
+
+# Full Documentation Generation - CLI Mode (/memory:docs-full-cli)
+
+## Overview
+
+Orchestrates project-wide documentation generation using CLI-based execution with batched agents and automatic tool fallback.
+
+**Parameters**:
+- `path`: Target directory (default: current directory)
+- `--tool <gemini|qwen|codex>`: Primary tool (default: gemini)
+
+**Execution Flow**: Discovery → Plan Presentation → Execution → Verification
+
+## 3-Layer Architecture & Auto-Strategy Selection
+
+### Layer Definition & Strategy Assignment
+
+| Layer | Depth | Strategy | Purpose | Context Pattern |
+|-------|-------|----------|---------|----------------|
+| **Layer 3** (Deepest) | ≥3 | `full` | Generate docs for all subdirectories with code | `@**/*` (all files) |
+| **Layer 2** (Middle) | 1-2 | `single` | Current dir + child docs | `@*/API.md @*/README.md @*.{ts,tsx,js,...}` |
+| **Layer 1** (Top) | 0 | `single` | Current dir + child docs | `@*/API.md @*/README.md @*.{ts,tsx,js,...}` |
+
+**Generation Direction**: Layer 3 → Layer 2 → Layer 1 (bottom-up dependency flow)
+
+**Strategy Auto-Selection**: Strategies are automatically determined by directory depth - no user configuration needed.
+
+### Strategy Details
+
+#### Full Strategy (Layer 3 Only)
+- **Use Case**: Deepest directories with comprehensive file coverage
+- **Behavior**: Generates API.md + README.md for current directory AND subdirectories containing code
+- **Context**: All files in current directory tree (`@**/*`)
+- **Output**: `.workflow/docs/{project_name}/{path}/API.md` + `README.md`
+
+
+#### Single Strategy (Layers 1-2)
+- **Use Case**: Upper layers that aggregate from existing documentation
+- **Behavior**: Generates API.md + README.md only in current directory
+- **Context**: Direct children docs + current directory code files
+- **Output**: `.workflow/docs/{project_name}/{path}/API.md` + `README.md`
+
+### Example Flow
+```
+src/auth/handlers/ (depth 3) → FULL STRATEGY
+  CONTEXT: @**/* (all files in handlers/ and subdirs)
+  GENERATES: .workflow/docs/project/src/auth/handlers/{API.md,README.md} + subdirs
+  ↓
+src/auth/ (depth 2) → SINGLE STRATEGY
+  CONTEXT: @*/API.md @*/README.md @*.ts (handlers docs + current code)
+  GENERATES: .workflow/docs/project/src/auth/{API.md,README.md} only
+  ↓
+src/ (depth 1) → SINGLE STRATEGY
+  CONTEXT: @*/API.md @*/README.md (auth docs, utils docs)
+  GENERATES: .workflow/docs/project/src/{API.md,README.md} only
+  ↓
+./ (depth 0) → SINGLE STRATEGY
+  CONTEXT: @*/API.md @*/README.md (src docs, tests docs)
+  GENERATES: .workflow/docs/project/{API.md,README.md} only
+```
+
+## Core Execution Rules
+
+1. **Analyze First**: Module discovery + folder classification before generation
+2. **Wait for Approval**: Present plan, no execution without user confirmation
+3. **Execution Strategy**:
+   - **<20 modules**: Direct parallel execution (max 4 concurrent per layer)
+   - **≥20 modules**: Agent batch processing (4 modules/agent, 73% overhead reduction)
+4. **Tool Fallback**: Auto-retry with fallback tools on failure
+5. **Layer Sequential**: Process layers 3→2→1 (bottom-up), parallel batches within layer
+6. **Safety Check**: Verify only docs files modified in .workflow/docs/
+7. **Layer-based Grouping**: Group modules by LAYER (not depth) for execution
+
+## Tool Fallback Hierarchy
+
+```javascript
+--tool gemini  →  [gemini, qwen, codex]  // default
+--tool qwen    →  [qwen, gemini, codex]
+--tool codex   →  [codex, gemini, qwen]
+```
+
+**Trigger**: Non-zero exit code from generation script
+
+| Tool   | Best For                       | Fallback To    |
+|--------|--------------------------------|----------------|
+| gemini | Documentation, patterns        | qwen → codex   |
+| qwen   | Architecture, system design    | gemini → codex |
+| codex  | Implementation, code quality   | gemini → qwen  |
+
+## Execution Phases
+
+### Phase 1: Discovery & Analysis
+
+```javascript
+// Get project metadata
+Bash({command: "pwd && basename \"$(pwd)\" && git rev-parse --show-toplevel 2>/dev/null || pwd", run_in_background: false});
+
+// Get module structure with classification
+Bash({command: "ccw tool exec get_modules_by_depth '{\"format\":\"list\"}' | ccw tool exec classify_folders '{}'", run_in_background: false});
+
+// OR with path parameter
+Bash({command: "cd <target-path> && ccw tool exec get_modules_by_depth '{\"format\":\"list\"}' | ccw tool exec classify_folders '{}'", run_in_background: false});
+```
+
+**Parse output** `depth:N|path:<PATH>|type:<code|navigation>|...` to extract module paths, types, and count.
+
+**Smart filter**: Auto-detect and skip tests/build/config/vendor based on project tech stack.
+
+### Phase 2: Plan Presentation
+
+**For <20 modules**:
+```
+Documentation Generation Plan:
+  Tool: gemini (fallback: qwen → codex)
+  Total: 7 modules
+  Execution: Direct parallel (< 20 modules threshold)
+  Project: myproject
+  Output: .workflow/docs/myproject/
+
+  Will generate docs for:
+  - ./core/interfaces (12 files, type: code) - depth 2 [Layer 2] - single strategy
+  - ./core (22 files, type: code) - depth 1 [Layer 2] - single strategy
+  - ./models (9 files, type: code) - depth 1 [Layer 2] - single strategy
+  - ./utils (12 files, type: navigation) - depth 1 [Layer 2] - single strategy
+  - . (5 files, type: code) - depth 0 [Layer 1] - single strategy
+
+  Documentation Strategy (Auto-Selected):
+  - Layer 2 (depth 1-2): API.md + README.md (current dir only, reference child docs)
+  - Layer 1 (depth 0): API.md + README.md (current dir only, reference child docs)
+
+  Output Structure:
+  - Code folders: API.md + README.md
+  - Navigation folders: README.md only
+
+  Auto-skipped: ./tests, __pycache__, node_modules (15 paths)
+  Execution order: Layer 2 → Layer 1
+  Estimated time: ~5-10 minutes
+
+  Confirm execution? (y/n)
+```
+
+**For ≥20 modules**:
+```
+Documentation Generation Plan:
+  Tool: gemini (fallback: qwen → codex)
+  Total: 31 modules
+  Execution: Agent batch processing (4 modules/agent)
+  Project: myproject
+  Output: .workflow/docs/myproject/
+
+  Will generate docs for:
+  - ./src/features/auth (12 files, type: code) - depth 3 [Layer 3] - full strategy
+  - ./.claude/commands/cli (6 files, type: code) - depth 3 [Layer 3] - full strategy
+  - ./src/utils (8 files, type: code) - depth 2 [Layer 2] - single strategy
+  ...
+
+  Documentation Strategy (Auto-Selected):
+  - Layer 3 (depth ≥3): API.md + README.md (all subdirs with code)
+  - Layer 2 (depth 1-2): API.md + README.md (current dir only)
+  - Layer 1 (depth 0): API.md + README.md (current dir only)
+
+  Output Structure:
+  - Code folders: API.md + README.md
+  - Navigation folders: README.md only
+
+  Auto-skipped: ./tests, __pycache__, node_modules (15 paths)
+  Execution order: Layer 3 → Layer 2 → Layer 1
+
+  Agent allocation (by LAYER):
+  - Layer 3 (14 modules, depth ≥3): 4 agents [4, 4, 4, 2]
+  - Layer 2 (15 modules, depth 1-2): 4 agents [4, 4, 4, 3]
+  - Layer 1 (2 modules, depth 0): 1 agent [2]
+
+  Estimated time: ~15-25 minutes
+
+  Confirm execution? (y/n)
+```
+
+### Phase 3A: Direct Execution (<20 modules)
+
+**Strategy**: Parallel execution within layer (max 4 concurrent), no agent overhead.
+
+**CRITICAL**: All Bash commands use `run_in_background: false` for synchronous execution.
+
+```javascript
+let project_name = detect_project_name();
+
+for (let layer of [3, 2, 1]) {
+  if (modules_by_layer[layer].length === 0) continue;
+  let batches = batch_modules(modules_by_layer[layer], 4);
+
+  for (let batch of batches) {
+    let parallel_tasks = batch.map(module => {
+      return async () => {
+        let strategy = module.depth >= 3 ? "full" : "single";
+        for (let tool of tool_order) {
+          Bash({
+            command: `cd ${module.path} && ccw tool exec generate_module_docs '{"strategy":"${strategy}","sourcePath":".","projectName":"${project_name}","tool":"${tool}"}'`,
+            run_in_background: false
+          });
+          if (bash_result.exit_code === 0) {
+            report(`✅ ${module.path} (Layer ${layer}) docs generated with ${tool}`);
+            return true;
+          }
+        }
+        report(`❌ FAILED: ${module.path} (Layer ${layer}) failed all tools`);
+        return false;
+      };
+    });
+    await Promise.all(parallel_tasks.map(task => task()));
+  }
+}
+```
+
+### Phase 3B: Agent Batch Execution (≥20 modules)
+
+**Strategy**: Batch modules into groups of 4, spawn memory-bridge agents per batch.
+
+```javascript
+// Group modules by LAYER and batch within each layer
+let modules_by_layer = group_by_layer(module_list);
+let tool_order = construct_tool_order(primary_tool);
+let project_name = detect_project_name();
+
+for (let layer of [3, 2, 1]) {
+  if (modules_by_layer[layer].length === 0) continue;
+
+  let batches = batch_modules(modules_by_layer[layer], 4);
+  let worker_tasks = [];
+
+  for (let batch of batches) {
+    worker_tasks.push(
+      Task(
+        subagent_type="memory-bridge",
+        description=`Generate docs for ${batch.length} modules in Layer ${layer}`,
+        prompt=generate_batch_worker_prompt(batch, tool_order, layer, project_name)
+      )
+    );
+  }
+
+  await parallel_execute(worker_tasks);
+}
+```
+
+**Batch Worker Prompt Template**:
+```
+PURPOSE: Generate documentation for assigned modules with tool fallback
+
+TASK: Generate API.md + README.md for assigned modules using specified strategies.
+
+PROJECT: {{project_name}}
+OUTPUT: .workflow/docs/{{project_name}}/
+
+MODULES:
+{{module_path_1}} (strategy: {{strategy_1}}, type: {{folder_type_1}})
+{{module_path_2}} (strategy: {{strategy_2}}, type: {{folder_type_2}})
+...
+
+TOOLS (try in order): {{tool_1}}, {{tool_2}}, {{tool_3}}
+
+EXECUTION SCRIPT: ccw tool exec generate_module_docs
+  - Accepts strategy parameter: full | single
+  - Accepts folder type detection: code | navigation
+  - Tool execution via direct CLI commands (gemini/qwen/codex)
+  - Output path: .workflow/docs/{{project_name}}/{module_path}/
+
+EXECUTION FLOW (for each module):
+  1. Tool fallback loop (exit on first success):
+     for tool in {{tool_1}} {{tool_2}} {{tool_3}}; do
+       Bash({
+         command: `cd "{{module_path}}" && ccw tool exec generate_module_docs '{"strategy":"{{strategy}}","sourcePath":".","projectName":"{{project_name}}","tool":"${tool}"}'`,
+         run_in_background: false
+       })
+       exit_code=$?
+
+       if [ $exit_code -eq 0 ]; then
+         report "✅ {{module_path}} docs generated with $tool"
+         break
+       else
+         report "⚠️  {{module_path}} failed with $tool, trying next..."
+         continue
+       fi
+     done
+
+  2. Handle complete failure (all tools failed):
+     if [ $exit_code -ne 0 ]; then
+       report "❌ FAILED: {{module_path}} - all tools exhausted"
+       # Continue to next module (do not abort batch)
+     fi
+
+FOLDER TYPE HANDLING:
+  - code: Generate API.md + README.md
+  - navigation: Generate README.md only
+
+FAILURE HANDLING:
+  - Module-level isolation: One module's failure does not affect others
+  - Exit code detection: Non-zero exit code triggers next tool
+  - Exhaustion reporting: Log modules where all tools failed
+  - Batch continuation: Always process remaining modules
+
+REPORTING FORMAT:
+  Per-module status:
+    ✅ path/to/module docs generated with {tool}
+    ⚠️  path/to/module failed with {tool}, trying next...
+    ❌ FAILED: path/to/module - all tools exhausted
+```
+
+### Phase 4: Project-Level Documentation
+
+**After all module documentation is generated, create project-level documentation files.**
+
+```javascript
+let project_name = detect_project_name();
+let project_root = get_project_root();
+
+// Step 1: Generate Project README
+report("Generating project README.md...");
+for (let tool of tool_order) {
+  Bash({
+    command: `cd ${project_root} && ccw tool exec generate_module_docs '{"strategy":"project-readme","sourcePath":".","projectName":"${project_name}","tool":"${tool}"}'`,
+    run_in_background: false
+  });
+  if (bash_result.exit_code === 0) {
+    report(`✅ Project README generated with ${tool}`);
+    break;
+  }
+}
+
+// Step 2: Generate Architecture & Examples
+report("Generating ARCHITECTURE.md and EXAMPLES.md...");
+for (let tool of tool_order) {
+  Bash({
+    command: `cd ${project_root} && ccw tool exec generate_module_docs '{"strategy":"project-architecture","sourcePath":".","projectName":"${project_name}","tool":"${tool}"}'`,
+    run_in_background: false
+  });
+  if (bash_result.exit_code === 0) {
+    report(`✅ Architecture docs generated with ${tool}`);
+    break;
+  }
+}
+
+// Step 3: Generate HTTP API documentation (if API routes detected)
+Bash({command: 'rg "router\\.|@Get|@Post" -g "*.{ts,js,py}" 2>/dev/null && echo "API_FOUND" || echo "NO_API"', run_in_background: false});
+if (bash_result.stdout.includes("API_FOUND")) {
+  report("Generating HTTP API documentation...");
+  for (let tool of tool_order) {
+    Bash({
+      command: `cd ${project_root} && ccw tool exec generate_module_docs '{"strategy":"http-api","sourcePath":".","projectName":"${project_name}","tool":"${tool}"}'`,
+      run_in_background: false
+    });
+    if (bash_result.exit_code === 0) {
+      report(`✅ HTTP API docs generated with ${tool}`);
+      break;
+    }
+  }
+}
+```
+
+**Expected Output**:
+```
+Project-Level Documentation:
+  ✅ README.md (project root overview)
+  ✅ ARCHITECTURE.md (system design)
+  ✅ EXAMPLES.md (usage examples)
+  ✅ api/README.md (HTTP API reference) [optional]
+```
+
+### Phase 5: Verification
+
+```javascript
+// Check documentation files created
+Bash({command: 'find .workflow/docs -type f -name "*.md" 2>/dev/null | wc -l', run_in_background: false});
+
+// Display structure
+Bash({command: 'tree -L 3 .workflow/docs/', run_in_background: false});
+```
+
+**Result Summary**:
+```
+Documentation Generation Summary:
+  Total: 31 | Success: 29 | Failed: 2
+  Tool usage: gemini: 25, qwen: 4, codex: 0
+  Failed: path1, path2
+
+  Generated documentation:
+    .workflow/docs/myproject/
+    ├── src/
+    │   ├── auth/
+    │   │   ├── API.md
+    │   │   └── README.md
+    │   └── utils/
+    │       └── README.md
+    └── README.md
+```
+
+## Error Handling
+
+**Batch Worker**: Tool fallback per module, batch isolation, clear status reporting
+**Coordinator**: Invalid path abort, user decline handling, verification with cleanup
+**Fallback Triggers**: Non-zero exit code, script timeout, unexpected output
+
+## Output Structure
+
+```
+.workflow/docs/{project_name}/
+├── src/                           # Mirrors source structure
+│   ├── modules/
+│   │   ├── README.md              # Navigation
+│   │   ├── auth/
+│   │   │   ├── API.md             # API signatures
+│   │   │   ├── README.md          # Module docs
+│   │   │   └── middleware/
+│   │   │       ├── API.md
+│   │   │       └── README.md
+│   │   └── api/
+│   │       ├── API.md
+│   │       └── README.md
+│   └── utils/
+│       └── README.md
+├── lib/
+│   └── core/
+│       ├── API.md
+│       └── README.md
+├── README.md                      # ✨ Project root overview (auto-generated)
+├── ARCHITECTURE.md                # ✨ System design (auto-generated)
+├── EXAMPLES.md                    # ✨ Usage examples (auto-generated)
+└── api/                           # ✨ Optional (auto-generated if HTTP API detected)
+    └── README.md                  # HTTP API reference
+```
+
+## Usage Examples
+
+```bash
+# Full project documentation generation
+/memory:docs-full-cli
+
+# Target specific directory
+/memory:docs-full-cli src/features/auth
+/memory:docs-full-cli .claude
+
+# Use specific tool
+/memory:docs-full-cli --tool qwen
+/memory:docs-full-cli src --tool qwen
+```
+
+## Key Advantages
+
+- **Efficiency**: 30 modules → 8 agents (73% reduction from sequential)
+- **Resilience**: 3-tier tool fallback per module
+- **Performance**: Parallel batches, no concurrency limits
+- **Observability**: Per-module tool usage, batch-level metrics
+- **Automation**: Zero configuration - strategy auto-selected by directory depth
+- **Path Mirroring**: Clear 1:1 mapping between source and documentation structure
+
+## Template Reference
+
+Templates used from `~/.claude/workflows/cli-templates/prompts/documentation/`:
+- `api.txt`: Code API documentation (Part A: Code API, Part B: HTTP API)
+- `module-readme.txt`: Module purpose, usage, dependencies
+- `folder-navigation.txt`: Navigation README for folders with subdirectories
+
+## Related Commands
+
+- `/memory:docs` - Agent-based documentation planning workflow
+- `/memory:docs-related-cli` - Update docs for changed modules only
+- `/workflow:execute` - Execute documentation tasks (when using agent mode)

.claude/commands/memory/docs-related-cli.md

@@ -0,0 +1,386 @@
+---
+name: docs-related-cli
+description: Generate/update documentation for git-changed modules using CLI execution with batched agents (4 modules/agent) and gemini→qwen→codex fallback, <15 modules uses direct parallel
+argument-hint: "[--tool <gemini|qwen|codex>]"
+---
+
+# Related Documentation Generation - CLI Mode (/memory:docs-related-cli)
+
+## Overview
+
+Orchestrates context-aware documentation generation/update for changed modules using CLI-based execution with batched agents and automatic tool fallback (gemini→qwen→codex).
+
+**Parameters**:
+- `--tool <gemini|qwen|codex>`: Primary tool (default: gemini)
+
+**Execution Flow**:
+1. Change Detection → 2. Plan Presentation → 3. Batched Execution → 4. Verification
+
+## Core Rules
+
+1. **Detect Changes First**: Use git diff to identify affected modules
+2. **Wait for Approval**: Present plan, no execution without user confirmation
+3. **Execution Strategy**:
+   - **<15 modules**: Direct parallel execution (max 4 concurrent per depth, no agent overhead)
+   - **≥15 modules**: Agent batch processing (4 modules/agent, 73% overhead reduction)
+4. **Tool Fallback**: Auto-retry with fallback tools on failure
+5. **Depth Sequential**: Process depths N→0, parallel batches within depth (both modes)
+6. **Related Mode**: Generate/update only changed modules and their parent contexts
+7. **Single Strategy**: Always use `single` strategy (incremental update)
+
+## Tool Fallback Hierarchy
+
+```javascript
+--tool gemini  →  [gemini, qwen, codex]  // default
+--tool qwen    →  [qwen, gemini, codex]
+--tool codex   →  [codex, gemini, qwen]
+```
+
+**Trigger**: Non-zero exit code from generation script
+
+| Tool   | Best For                       | Fallback To    |
+|--------|--------------------------------|----------------|
+| gemini | Documentation, patterns        | qwen → codex   |
+| qwen   | Architecture, system design    | gemini → codex |
+| codex  | Implementation, code quality   | gemini → qwen  |
+
+## Phase 1: Change Detection & Analysis
+
+```javascript
+// Get project metadata
+Bash({command: "pwd && basename \"$(pwd)\" && git rev-parse --show-toplevel 2>/dev/null || pwd", run_in_background: false});
+
+// Detect changed modules
+Bash({command: "ccw tool exec detect_changed_modules '{\"format\":\"list\"}'", run_in_background: false});
+
+// Cache git changes
+Bash({command: "git add -A 2>/dev/null || true", run_in_background: false});
+```
+
+**Parse output** `depth:N|path:<PATH>|change:<TYPE>|type:<code|navigation>` to extract affected modules.
+
+**Smart filter**: Auto-detect and skip tests/build/config/vendor based on project tech stack (Node.js/Python/Go/Rust/etc).
+
+**Fallback**: If no changes detected, use recent modules (first 10 by depth).
+
+## Phase 2: Plan Presentation
+
+**Present filtered plan**:
+```
+Related Documentation Generation Plan:
+  Tool: gemini (fallback: qwen → codex)
+  Changed: 4 modules | Batching: 4 modules/agent
+  Project: myproject
+  Output: .workflow/docs/myproject/
+
+  Will generate/update docs for:
+  - ./src/api/auth (5 files, type: code) [new module]
+  - ./src/api (12 files, type: code) [parent of changed auth/]
+  - ./src (8 files, type: code) [parent context]
+  - . (14 files, type: code) [root level]
+
+  Documentation Strategy:
+  - Strategy: single (all modules - incremental update)
+  - Output: API.md + README.md (code folders), README.md only (navigation folders)
+  - Context: Current dir code + child docs
+
+  Auto-skipped (12 paths):
+  - Tests: ./src/api/auth.test.ts (8 paths)
+  - Config: tsconfig.json (3 paths)
+  - Other: node_modules (1 path)
+
+  Agent allocation:
+  - Depth 3 (1 module): 1 agent [1]
+  - Depth 2 (1 module): 1 agent [1]
+  - Depth 1 (1 module): 1 agent [1]
+  - Depth 0 (1 module): 1 agent [1]
+
+  Estimated time: ~5-10 minutes
+
+  Confirm execution? (y/n)
+```
+
+**Decision logic**:
+- User confirms "y": Proceed with execution
+- User declines "n": Abort, no changes
+- <15 modules: Direct execution
+- ≥15 modules: Agent batch execution
+
+## Phase 3A: Direct Execution (<15 modules)
+
+**Strategy**: Parallel execution within depth (max 4 concurrent), no agent overhead.
+
+**CRITICAL**: All Bash commands use `run_in_background: false` for synchronous execution.
+
+```javascript
+let project_name = detect_project_name();
+
+for (let depth of sorted_depths.reverse()) {  // N → 0
+  let batches = batch_modules(modules_by_depth[depth], 4);
+
+  for (let batch of batches) {
+    let parallel_tasks = batch.map(module => {
+      return async () => {
+        for (let tool of tool_order) {
+          Bash({
+            command: `cd ${module.path} && ccw tool exec generate_module_docs '{"strategy":"single","sourcePath":".","projectName":"${project_name}","tool":"${tool}"}'`,
+            run_in_background: false
+          });
+          if (bash_result.exit_code === 0) {
+            report(`✅ ${module.path} docs generated with ${tool}`);
+            return true;
+          }
+        }
+        report(`❌ FAILED: ${module.path} failed all tools`);
+        return false;
+      };
+    });
+    await Promise.all(parallel_tasks.map(task => task()));
+  }
+}
+```
+
+## Phase 3B: Agent Batch Execution (≥15 modules)
+
+### Batching Strategy
+
+```javascript
+// Batch modules into groups of 4
+function batch_modules(modules, batch_size = 4) {
+  let batches = [];
+  for (let i = 0; i < modules.length; i += batch_size) {
+    batches.push(modules.slice(i, i + batch_size));
+  }
+  return batches;
+}
+// Examples: 10→[4,4,2] | 8→[4,4] | 3→[3]
+```
+
+### Coordinator Orchestration
+
+```javascript
+let modules_by_depth = group_by_depth(changed_modules);
+let tool_order = construct_tool_order(primary_tool);
+let project_name = detect_project_name();
+
+for (let depth of sorted_depths.reverse()) {  // N → 0
+  let batches = batch_modules(modules_by_depth[depth], 4);
+  let worker_tasks = [];
+
+  for (let batch of batches) {
+    worker_tasks.push(
+      Task(
+        subagent_type="memory-bridge",
+        description=`Generate docs for ${batch.length} modules at depth ${depth}`,
+        prompt=generate_batch_worker_prompt(batch, tool_order, depth, project_name, "related")
+      )
+    );
+  }
+
+  await parallel_execute(worker_tasks);  // Batches run in parallel
+}
+```
+
+### Batch Worker Prompt Template
+
+```
+PURPOSE: Generate/update documentation for assigned modules with tool fallback (related mode)
+
+TASK:
+Generate documentation for the following modules based on recent changes. For each module, try tools in order until success.
+
+PROJECT: {{project_name}}
+OUTPUT: .workflow/docs/{{project_name}}/
+
+MODULES:
+{{module_path_1}} (type: {{folder_type_1}})
+{{module_path_2}} (type: {{folder_type_2}})
+{{module_path_3}} (type: {{folder_type_3}})
+{{module_path_4}} (type: {{folder_type_4}})
+
+TOOLS (try in order):
+1. {{tool_1}}
+2. {{tool_2}}
+3. {{tool_3}}
+
+EXECUTION:
+For each module above:
+  1. Try tool 1:
+     Bash({
+       command: `cd "{{module_path}}" && ccw tool exec generate_module_docs '{"strategy":"single","sourcePath":".","projectName":"{{project_name}}","tool":"{{tool_1}}"}'`,
+       run_in_background: false
+     })
+     → Success: Report "✅ {{module_path}} docs generated with {{tool_1}}", proceed to next module
+     → Failure: Try tool 2
+  2. Try tool 2:
+     Bash({
+       command: `cd "{{module_path}}" && ccw tool exec generate_module_docs '{"strategy":"single","sourcePath":".","projectName":"{{project_name}}","tool":"{{tool_2}}"}'`,
+       run_in_background: false
+     })
+     → Success: Report "✅ {{module_path}} docs generated with {{tool_2}}", proceed to next module
+     → Failure: Try tool 3
+  3. Try tool 3:
+     Bash({
+       command: `cd "{{module_path}}" && ccw tool exec generate_module_docs '{"strategy":"single","sourcePath":".","projectName":"{{project_name}}","tool":"{{tool_3}}"}'`,
+       run_in_background: false
+     })
+     → Success: Report "✅ {{module_path}} docs generated with {{tool_3}}", proceed to next module
+     → Failure: Report "❌ FAILED: {{module_path}} failed all tools", proceed to next module
+
+FOLDER TYPE HANDLING:
+  - code: Generate API.md + README.md
+  - navigation: Generate README.md only
+
+REPORTING:
+Report final summary with:
+- Total processed: X modules
+- Successful: Y modules
+- Failed: Z modules
+- Tool usage: {{tool_1}}:X, {{tool_2}}:Y, {{tool_3}}:Z
+```
+
+## Phase 4: Verification
+
+```javascript
+// Check documentation files created/updated
+Bash({command: 'find .workflow/docs -type f -name "*.md" 2>/dev/null | wc -l', run_in_background: false});
+
+// Display recent changes
+Bash({command: 'find .workflow/docs -type f -name "*.md" -mmin -60 2>/dev/null', run_in_background: false});
+```
+
+**Aggregate results**:
+```
+Documentation Generation Summary:
+  Total: 4 | Success: 4 | Failed: 0
+
+  Tool usage:
+  - gemini: 4 modules
+  - qwen: 0 modules (fallback)
+  - codex: 0 modules
+
+  Changes:
+  .workflow/docs/myproject/src/api/auth/API.md      (new)
+  .workflow/docs/myproject/src/api/auth/README.md   (new)
+  .workflow/docs/myproject/src/api/API.md           (updated)
+  .workflow/docs/myproject/src/api/README.md        (updated)
+  .workflow/docs/myproject/src/API.md               (updated)
+  .workflow/docs/myproject/src/README.md            (updated)
+  .workflow/docs/myproject/API.md                   (updated)
+  .workflow/docs/myproject/README.md                (updated)
+```
+
+## Execution Summary
+
+**Module Count Threshold**:
+- **<15 modules**: Coordinator executes Phase 3A (Direct Execution)
+- **≥15 modules**: Coordinator executes Phase 3B (Agent Batch Execution)
+
+**Agent Hierarchy** (for ≥15 modules):
+- **Coordinator**: Handles batch division, spawns worker agents per depth
+- **Worker Agents**: Each processes 4 modules with tool fallback (related mode)
+
+## Error Handling
+
+**Batch Worker**:
+- Tool fallback per module (auto-retry)
+- Batch isolation (failures don't propagate)
+- Clear per-module status reporting
+
+**Coordinator**:
+- No changes: Use fallback (recent 10 modules)
+- User decline: No execution
+- Verification fail: Report incomplete modules
+- Partial failures: Continue execution, report failed modules
+
+**Fallback Triggers**:
+- Non-zero exit code
+- Script timeout
+- Unexpected output
+
+## Output Structure
+
+```
+.workflow/docs/{project_name}/
+├── src/                           # Mirrors source structure
+│   ├── modules/
+│   │   ├── README.md
+│   │   ├── auth/
+│   │   │   ├── API.md             # Updated based on code changes
+│   │   │   └── README.md          # Updated based on code changes
+│   │   └── api/
+│   │       ├── API.md
+│   │       └── README.md
+│   └── utils/
+│       └── README.md
+└── README.md
+```
+
+## Usage Examples
+
+```bash
+# Daily development documentation update
+/memory:docs-related-cli
+
+# After feature work with specific tool
+/memory:docs-related-cli --tool qwen
+
+# Code quality documentation review after implementation
+/memory:docs-related-cli --tool codex
+```
+
+## Key Advantages
+
+**Efficiency**: 30 modules → 8 agents (73% reduction)
+**Resilience**: 3-tier fallback per module
+**Performance**: Parallel batches, no concurrency limits
+**Context-aware**: Updates based on actual git changes
+**Fast**: Only affected modules, not entire project
+**Incremental**: Single strategy for focused updates
+
+## Coordinator Checklist
+
+- Parse `--tool` (default: gemini)
+- Get project metadata (name, root)
+- Detect changed modules via detect_changed_modules.sh
+- **Smart filter modules** (auto-detect tech stack, skip tests/build/config/vendor)
+- Cache git changes
+- Apply fallback if no changes (recent 10 modules)
+- Construct tool fallback order
+- **Present filtered plan** with skip reasons and change types
+- **Wait for y/n confirmation**
+- Determine execution mode:
+  - **<15 modules**: Direct execution (Phase 3A)
+    - For each depth (N→0): Sequential module updates with tool fallback
+  - **≥15 modules**: Agent batch execution (Phase 3B)
+    - For each depth (N→0): Batch modules (4 per batch), spawn batch workers in parallel
+- Wait for depth/batch completion
+- Aggregate results
+- Verification check (documentation files created/updated)
+- Display summary + recent changes
+
+## Comparison with Full Documentation Generation
+
+| Aspect | Related Generation | Full Generation |
+|--------|-------------------|-----------------|
+| **Scope** | Changed modules only | All project modules |
+| **Speed** | Fast (minutes) | Slower (10-30 min) |
+| **Use case** | Daily development | Initial setup, major refactoring |
+| **Strategy** | `single` (all) | `full` (L3) + `single` (L1-2) |
+| **Trigger** | After commits | After setup or major changes |
+| **Batching** | 4 modules/agent | 4 modules/agent |
+| **Fallback** | gemini→qwen→codex | gemini→qwen→codex |
+| **Complexity threshold** | ≤15 modules | ≤20 modules |
+
+## Template Reference
+
+Templates used from `~/.claude/workflows/cli-templates/prompts/documentation/`:
+- `api.txt`: Code API documentation
+- `module-readme.txt`: Module purpose, usage, dependencies
+- `folder-navigation.txt`: Navigation README for folders
+
+## Related Commands
+
+- `/memory:docs-full-cli` - Full project documentation generation
+- `/memory:docs` - Agent-based documentation planning workflow
+- `/memory:update-related` - Update CLAUDE.md for changed modules

.claude/commands/memory/docs.md

@@ -1,6 +1,6 @@
 ---
 name: docs
-description: Documentation planning and orchestration - creates structured documentation tasks for execution
+description: Plan documentation workflow with dynamic grouping (≤10 docs/task), generates IMPL tasks for parallel module trees, README, ARCHITECTURE, and HTTP API docs
 argument-hint: "[path] [--tool <gemini|qwen|codex>] [--mode <full|partial>] [--cli-execute]"
 ---
 
@@ -36,7 +36,6 @@ Lightweight planner that analyzes project structure, decomposes documentation wo
 | `my_app/src/modules/auth/` | `my_app` | `.workflow/docs/my_app/src/modules/auth/API.md` |
 | `another_project/lib/utils/` | `another_project` | `.workflow/docs/another_project/lib/utils/API.md` |
 
-**Benefits**: Easy to locate documentation, maintains logical organization, clear 1:1 mapping, supports any project structure.
 
 ## Parameters
 
@@ -44,7 +43,11 @@ Lightweight planner that analyzes project structure, decomposes documentation wo
 /memory:docs [path] [--tool <gemini|qwen|codex>] [--mode <full|partial>] [--cli-execute]
 ```
 
-- **path**: Target directory (default: current directory)
+- **path**: Source directory to analyze (default: current directory)
+  - Specifies the source code directory to be documented
+  - Documentation is generated in a separate `.workflow/docs/{project_name}/` directory at the workspace root, **not** within the source `path` itself
+  - The source path's structure is mirrored within the project-specific documentation folder
+  - Example: analyzing `src/modules` produces documentation at `.workflow/docs/{project_name}/src/modules/`
 - **--mode**: Documentation generation mode (default: full)
   - `full`: Complete documentation (modules + README + ARCHITECTURE + EXAMPLES + HTTP API)
   - `partial`: Module documentation only (API.md + README.md)
@@ -61,12 +64,17 @@ Lightweight planner that analyzes project structure, decomposes documentation wo
 ```bash
 # Get target path, project name, and root
 bash(pwd && basename "$(pwd)" && git rev-parse --show-toplevel 2>/dev/null || pwd && date +%Y%m%d-%H%M%S)
+```
 
-# Create session directories (replace timestamp)
-bash(mkdir -p .workflow/WFS-docs-{timestamp}/.{task,process,summaries} && touch .workflow/.active-WFS-docs-{timestamp})
+```javascript
+// Create docs session (type: docs)
+SlashCommand(command="/workflow:session:start --type docs --new \"{project_name}-docs-{timestamp}\"")
+// Parse output to get sessionId
+```
 
-# Create workflow-session.json (replace values)
-bash(echo '{"session_id":"WFS-docs-{timestamp}","project":"{project} documentation","status":"planning","timestamp":"2024-01-20T14:30:22+08:00","path":".","target_path":"{target_path}","project_root":"{project_root}","project_name":"{project_name}","mode":"full","tool":"gemini","cli_execute":false}' | jq '.' > .workflow/WFS-docs-{timestamp}/workflow-session.json)
+```bash
+# Update workflow-session.json with docs-specific fields
+bash(jq '. + {"target_path":"{target_path}","project_root":"{project_root}","project_name":"{project_name}","mode":"full","tool":"gemini","cli_execute":false}' .workflow/active/{sessionId}/workflow-session.json > tmp.json && mv tmp.json .workflow/active/{sessionId}/workflow-session.json)
 ```
 
 ### Phase 2: Analyze Structure
@@ -77,10 +85,10 @@ bash(echo '{"session_id":"WFS-docs-{timestamp}","project":"{project} documentati
 
 ```bash
 # 1. Run folder analysis
-bash(~/.claude/scripts/get_modules_by_depth.sh | ~/.claude/scripts/classify-folders.sh)
+bash(ccw tool exec get_modules_by_depth '{}' | ccw tool exec classify_folders '{}')
 
 # 2. Get top-level directories (first 2 path levels)
-bash(~/.claude/scripts/get_modules_by_depth.sh | ~/.claude/scripts/classify-folders.sh | awk -F'|' '{print $1}' | sed 's|^\./||' | awk -F'/' '{if(NF>=2) print $1"/"$2; else if(NF==1) print $1}' | sort -u)
+bash(ccw tool exec get_modules_by_depth '{}' | ccw tool exec classify_folders '{}' | awk -F'|' '{print $1}' | sed 's|^\./||' | awk -F'/' '{if(NF>=2) print $1"/"$2; else if(NF==1) print $1}' | sort -u)
 
 # 3. Find existing docs (if directory exists)
 bash(if [ -d .workflow/docs/\${project_name} ]; then find .workflow/docs/\${project_name} -type f -name "*.md" ! -path "*/README.md" ! -path "*/ARCHITECTURE.md" ! -path "*/EXAMPLES.md" ! -path "*/api/*" 2>/dev/null; fi)
@@ -89,7 +97,7 @@ bash(if [ -d .workflow/docs/\${project_name} ]; then find .workflow/docs/\${proj
 bash(if [ -d .workflow/docs/\${project_name} ]; then find .workflow/docs/\${project_name} -type f -name "*.md" ! -path "*/README.md" ! -path "*/ARCHITECTURE.md" ! -path "*/EXAMPLES.md" ! -path "*/api/*" 2>/dev/null | xargs cat 2>/dev/null; fi)
 ```
 
-**Data Processing**: Parse bash outputs, calculate statistics, use **Write tool** to create `${session_dir}/.process/phase2-analysis.json` with structure:
+**Data Processing**: Parse bash outputs, calculate statistics, use **Write tool** to create `${session_dir}/.process/doc-planning-data.json` with structure:
 
 ```json
 {
@@ -118,7 +126,7 @@ bash(if [ -d .workflow/docs/\${project_name} ]; then find .workflow/docs/\${proj
 
 **Then** use **Edit tool** to update `workflow-session.json` adding analysis field.
 
-**Output**: Single `phase2-analysis.json` with all analysis data (no temp files or Python scripts).
+**Output**: Single `doc-planning-data.json` with all analysis data (no temp files or Python scripts).
 
 **Auto-skipped**: Tests (`**/test/**`, `**/*.test.*`), Build (`**/node_modules/**`, `**/dist/**`), Config (root-level files), Vendor directories.
 
@@ -127,8 +135,8 @@ bash(if [ -d .workflow/docs/\${project_name} ]; then find .workflow/docs/\${proj
 **Commands**:
 
 ```bash
-# Count existing docs from phase2-analysis.json
-bash(cat .workflow/WFS-docs-{timestamp}/.process/phase2-analysis.json | jq '.existing_docs.file_list | length')
+# Count existing docs from doc-planning-data.json
+bash(cat .workflow/active/WFS-docs-{timestamp}/.process/doc-planning-data.json | jq '.existing_docs.file_list | length')
 ```
 
 **Data Processing**: Use count result, then use **Edit tool** to update `workflow-session.json`:
@@ -177,16 +185,15 @@ Large Projects (single dir >10 docs):
 4. If single dir exceeds 10 docs, split by subdirectories
 5. Create parallel Level 1 tasks with ≤10 docs each
 
-**Benefits**: Parallel execution, failure isolation, progress visibility, context sharing, document count control.
 
 **Commands**:
 
 ```bash
-# 1. Get top-level directories from phase2-analysis.json
-bash(cat .workflow/WFS-docs-{timestamp}/.process/phase2-analysis.json | jq -r '.top_level_dirs[]')
+# 1. Get top-level directories from doc-planning-data.json
+bash(cat .workflow/active/WFS-docs-{timestamp}/.process/doc-planning-data.json | jq -r '.top_level_dirs[]')
 
 # 2. Get mode from workflow-session.json
-bash(cat .workflow/WFS-docs-{timestamp}/workflow-session.json | jq -r '.mode // "full"')
+bash(cat .workflow/active/WFS-docs-{timestamp}/workflow-session.json | jq -r '.mode // "full"')
 
 # 3. Check for HTTP API
 bash(grep -r "router\.|@Get\|@Post" src/ 2>/dev/null && echo "API_FOUND" || echo "NO_API")
@@ -201,7 +208,7 @@ bash(grep -r "router\.|@Get\|@Post" src/ 2>/dev/null && echo "API_FOUND" || echo
    - If total ≤10 docs: create group
    - If total >10 docs: split to 1 dir/group or subdivide
    - If single dir >10 docs: split by subdirectories
-3. Use **Edit tool** to update `phase2-analysis.json` adding groups field:
+3. Use **Edit tool** to update `doc-planning-data.json` adding groups field:
    ```json
    "groups": {
      "count": 3,
@@ -215,7 +222,7 @@ bash(grep -r "router\.|@Get\|@Post" src/ 2>/dev/null && echo "API_FOUND" || echo
 
 **Task ID Calculation**:
 ```bash
-group_count=$(jq '.groups.count' .workflow/WFS-docs-{timestamp}/.process/phase2-analysis.json)
+group_count=$(jq '.groups.count' .workflow/active/WFS-docs-{timestamp}/.process/doc-planning-data.json)
 readme_id=$((group_count + 1))   # Next ID after groups
 arch_id=$((group_count + 2))
 api_id=$((group_count + 3))
@@ -228,16 +235,16 @@ api_id=$((group_count + 3))
 | Mode | cli_execute | Placement | CLI MODE | Approval Flag | Agent Role |
 |------|-------------|-----------|----------|---------------|------------|
 | **Agent** | false | pre_analysis | analysis | (none) | Generate docs in implementation_approach |
-| **CLI** | true | implementation_approach | write | --approval-mode yolo | Execute CLI commands, validate output |
+| **CLI** | true | implementation_approach | write | --mode write | Execute CLI commands, validate output |
 
 **Command Patterns**:
-- Gemini/Qwen: `cd dir && gemini -p "..."`
-- CLI Mode: `cd dir && gemini --approval-mode yolo -p "..."`
-- Codex: `codex -C dir --full-auto exec "..." --skip-git-repo-check -s danger-full-access`
+- Gemini/Qwen: `ccw cli -p "..." --tool gemini --mode analysis --cd dir`
+- CLI Mode: `ccw cli -p "..." --tool gemini --mode write --cd dir`
+- Codex: `ccw cli -p "..." --tool codex --mode write --cd dir`
 
 **Generation Process**:
 1. Read configuration values (tool, cli_execute, mode) from workflow-session.json
-2. Read group assignments from phase2-analysis.json
+2. Read group assignments from doc-planning-data.json
 3. Generate Level 1 tasks (IMPL-001 to IMPL-N, one per group)
 4. Generate Level 2+ tasks if mode=full (README, ARCHITECTURE, HTTP API)
 
@@ -262,14 +269,14 @@ api_id=$((group_count + 3))
   },
   "context": {
     "requirements": [
-      "Process directories from group ${group_number} in phase2-analysis.json",
+      "Process directories from group ${group_number} in doc-planning-data.json",
       "Generate docs to .workflow/docs/${project_name}/ (mirrored structure)",
       "Code folders: API.md + README.md; Navigation folders: README.md only",
       "Use pre-analyzed data from Phase 2 (no redundant analysis)"
     ],
     "focus_paths": ["${group_dirs_from_json}"],
     "precomputed_data": {
-      "phase2_analysis": "${session_dir}/.process/phase2-analysis.json"
+      "phase2_analysis": "${session_dir}/.process/doc-planning-data.json"
     }
   },
   "flow_control": {
@@ -278,8 +285,8 @@ api_id=$((group_count + 3))
         "step": "load_precomputed_data",
         "action": "Load Phase 2 analysis and extract group directories",
         "commands": [
-          "bash(cat ${session_dir}/.process/phase2-analysis.json)",
-          "bash(jq '.groups.assignments[] | select(.group_id == \"${group_number}\") | .directories' ${session_dir}/.process/phase2-analysis.json)"
+          "bash(cat ${session_dir}/.process/doc-planning-data.json)",
+          "bash(jq '.groups.assignments[] | select(.group_id == \"${group_number}\") | .directories' ${session_dir}/.process/doc-planning-data.json)"
         ],
         "output_to": "phase2_context",
         "note": "Single JSON file contains all Phase 2 analysis results"
@@ -324,7 +331,7 @@ api_id=$((group_count + 3))
 {
   "step": 2,
   "title": "Batch generate documentation via CLI",
-  "command": "bash(dirs=$(jq -r '.groups.assignments[] | select(.group_id == \"${group_number}\") | .directories[]' ${session_dir}/.process/phase2-analysis.json); for dir in $dirs; do cd \"$dir\" && gemini --approval-mode yolo -p \"PURPOSE: Generate module docs\\nTASK: Create documentation\\nMODE: write\\nCONTEXT: @**/* [phase2_context]\\nEXPECTED: API.md and README.md\\nRULES: Mirror structure\" || echo \"Failed: $dir\"; cd -; done)",
+  "command": "ccw cli -p 'PURPOSE: Generate module docs\\nTASK: Create documentation\\nMODE: write\\nCONTEXT: @**/* [phase2_context]\\nEXPECTED: API.md and README.md\\nRULES: Mirror structure' --tool gemini --mode write --cd ${dirs_from_group}",
   "depends_on": [1],
   "output": "generated_docs"
 }
@@ -356,7 +363,7 @@ api_id=$((group_count + 3))
       },
       {
         "step": "analyze_project",
-        "command": "bash(gemini \"PURPOSE: Analyze project structure\\nTASK: Extract overview from modules\\nMODE: analysis\\nCONTEXT: [all_module_docs]\\nEXPECTED: Project outline\")",
+        "command": "bash(ccw cli -p \"PURPOSE: Analyze project structure\\nTASK: Extract overview from modules\\nMODE: analysis\\nCONTEXT: [all_module_docs]\\nEXPECTED: Project outline\" --tool gemini --mode analysis)",
         "output_to": "project_outline"
       }
     ],
@@ -396,7 +403,7 @@ api_id=$((group_count + 3))
     "pre_analysis": [
       {"step": "load_existing_docs", "command": "bash(cat .workflow/docs/${project_name}/{ARCHITECTURE,EXAMPLES}.md 2>/dev/null || echo 'No existing docs')", "output_to": "existing_arch_examples"},
       {"step": "load_all_docs", "command": "bash(cat .workflow/docs/${project_name}/README.md && find .workflow/docs/${project_name} -type f -name '*.md' ! -path '*/README.md' ! -path '*/ARCHITECTURE.md' ! -path '*/EXAMPLES.md' ! -path '*/api/*' | xargs cat)", "output_to": "all_docs"},
-      {"step": "analyze_architecture", "command": "bash(gemini \"PURPOSE: Analyze system architecture\\nTASK: Synthesize architectural overview and examples\\nMODE: analysis\\nCONTEXT: [all_docs]\\nEXPECTED: Architecture + Examples outline\")", "output_to": "arch_examples_outline"}
+      {"step": "analyze_architecture", "command": "bash(ccw cli -p \"PURPOSE: Analyze system architecture\\nTASK: Synthesize architectural overview and examples\\nMODE: analysis\\nCONTEXT: [all_docs]\\nEXPECTED: Architecture + Examples outline\" --tool gemini --mode analysis)", "output_to": "arch_examples_outline"}
     ],
     "implementation_approach": [
       {
@@ -433,7 +440,7 @@ api_id=$((group_count + 3))
     "pre_analysis": [
       {"step": "discover_api", "command": "bash(rg 'router\\.| @(Get|Post)' -g '*.{ts,js}')", "output_to": "endpoint_discovery"},
       {"step": "load_existing_api", "command": "bash(cat .workflow/docs/${project_name}/api/README.md 2>/dev/null || echo 'No existing API docs')", "output_to": "existing_api_docs"},
-      {"step": "analyze_api", "command": "bash(gemini \"PURPOSE: Document HTTP API\\nTASK: Analyze endpoints\\nMODE: analysis\\nCONTEXT: @src/api/**/* [endpoint_discovery]\\nEXPECTED: API outline\")", "output_to": "api_outline"}
+      {"step": "analyze_api", "command": "bash(ccw cli -p \"PURPOSE: Document HTTP API\\nTASK: Analyze endpoints\\nMODE: analysis\\nCONTEXT: @src/api/**/* [endpoint_discovery]\\nEXPECTED: API outline\" --tool gemini --mode analysis)", "output_to": "api_outline"}
     ],
     "implementation_approach": [
       {
@@ -458,14 +465,13 @@ api_id=$((group_count + 3))
 **Unified Structure** (single JSON replaces multiple text files):
 
 ```
-.workflow/
-├── .active-WFS-docs-{timestamp}
+.workflow/active/
 └── WFS-docs-{timestamp}/
     ├── workflow-session.json            # Session metadata
     ├── IMPL_PLAN.md
     ├── TODO_LIST.md
     ├── .process/
-    │   └── phase2-analysis.json         # All Phase 2 analysis data (replaces 7+ files)
+    │   └── doc-planning-data.json       # All Phase 2 analysis data (replaces 7+ files)
     └── .task/
         ├── IMPL-001.json                # Small: all modules | Large: group 1
         ├── IMPL-00N.json                # (Large only: groups 2-N)
@@ -474,7 +480,7 @@ api_id=$((group_count + 3))
         └── IMPL-{N+3}.json              # HTTP API (optional)
 ```
 
-**phase2-analysis.json Structure**:
+**doc-planning-data.json Structure**:
 ```json
 {
   "metadata": {
@@ -595,7 +601,7 @@ api_id=$((group_count + 3))
 | Mode | CLI Placement | CLI MODE | Approval Flag | Agent Role |
 |------|---------------|----------|---------------|------------|
 | **Agent (default)** | pre_analysis | analysis | (none) | Generates documentation content |
-| **CLI (--cli-execute)** | implementation_approach | write | --approval-mode yolo | Executes CLI commands, validates output |
+| **CLI (--cli-execute)** | implementation_approach | write | --mode write | Executes CLI commands, validates output |
 
 **Execution Flow**:
 - **Phase 2**: Unified analysis once, results in `.process/`

.claude/commands/memory/load-skill-memory.md

@@ -1,6 +1,6 @@
 ---
 name: load-skill-memory
-description: Activate SKILL package (auto-detect or manual) and load documentation based on task intent
+description: Activate SKILL package (auto-detect from paths/keywords or manual) and intelligently load documentation based on task intent keywords
 argument-hint: "[skill_name] \"task intent description\""
 allowed-tools: Bash(*), Read(*), Skill(*)
 ---

.claude/commands/memory/load.md

@@ -1,11 +1,11 @@
 ---
 name: load
-description: Load project memory by delegating to agent, returns structured core content package for subsequent operations
+description: Delegate to universal-executor agent to analyze project via Gemini/Qwen CLI and return JSON core content package for task context
 argument-hint: "[--tool gemini|qwen] \"task context description\""
 allowed-tools: Task(*), Bash(*)
 examples:
   - /memory:load "在当前前端基础上开发用户认证功能"
-  - /memory:load --tool qwen -p "重构支付模块API"
+  - /memory:load --tool qwen "重构支付模块API"
 ---
 
 # Memory Load Command (/memory:load)
@@ -39,7 +39,7 @@ The command fully delegates to **universal-executor agent**, which autonomously:
 1. **Analyzes Project Structure**: Executes `get_modules_by_depth.sh` to understand architecture
 2. **Loads Documentation**: Reads CLAUDE.md, README.md and other key docs
 3. **Extracts Keywords**: Derives core keywords from task description
-4. **Discovers Files**: Uses MCP code-index or rg/find to locate relevant files
+4. **Discovers Files**: Uses CodexLens MCP or rg/find to locate relevant files
 5. **CLI Deep Analysis**: Executes Gemini/Qwen CLI for deep context analysis
 6. **Generates Content Package**: Returns structured JSON core content package
 
@@ -109,7 +109,7 @@ Task(
 
 1. **Project Structure**
    \`\`\`bash
-   bash(~/.claude/scripts/get_modules_by_depth.sh)
+   bash(ccw tool exec get_modules_by_depth '{}')
    \`\`\`
 
 2. **Core Documentation**
@@ -136,7 +136,7 @@ Task(
 Execute Gemini/Qwen CLI for deep analysis (saves main thread tokens):
 
 \`\`\`bash
-cd . && ${tool} -p "
+ccw cli -p "
 PURPOSE: Extract project core context for task: ${task_description}
 TASK: Analyze project architecture, tech stack, key patterns, relevant files
 MODE: analysis
@@ -147,7 +147,7 @@ RULES:
 - Identify key architecture patterns and technical constraints
 - Extract integration points and development standards
 - Output concise, structured format
-"
+" --tool ${tool} --mode analysis
 \`\`\`
 
 ### Step 4: Generate Core Content Package
@@ -212,7 +212,7 @@ Before returning:
 ### Example 2: Using Qwen Tool
 
 ```bash
-/memory:load --tool qwen -p "重构支付模块API"
+/memory:load --tool qwen "重构支付模块API"
 ```
 
 Agent uses Qwen CLI for analysis, returns same structured package.

.claude/commands/memory/skill-memory.md

@@ -1,6 +1,6 @@
 ---
 name: skill-memory
-description: Generate SKILL package index from project documentation
+description: 4-phase autonomous orchestrator: check docs → /memory:docs planning → /workflow:execute → generate SKILL.md with progressive loading index (skips phases 2-3 if docs exist)
 argument-hint: "[path] [--tool <gemini|qwen|codex>] [--regenerate] [--mode <full|partial>] [--cli-execute]"
 allowed-tools: SlashCommand(*), TodoWrite(*), Bash(*), Read(*), Write(*)
 ---
@@ -508,16 +508,7 @@ User triggers command
 
 ---
 
-## Benefits
 
-- **Pure Orchestrator**: No task JSON generation, delegates to /memory:docs
-- **Auto-Continue**: Autonomous 4-phase execution without user interaction
-- **Intelligent Skip**: Detects existing docs and skips regeneration for fast SKILL updates
-- **Always Fresh Index**: Phase 4 always executes to ensure SKILL.md stays synchronized
-- **Simplified**: ~70% less code than previous version
-- **Maintainable**: Changes to /memory:docs automatically apply
-- **Direct Generation**: Phase 4 directly writes SKILL.md
-- **Flexible**: Supports all /memory:docs options (tool, mode, cli-execute)
 
 ## Architecture
 

.claude/commands/memory/style-skill-memory.md

@@ -0,0 +1,396 @@
+---
+name: style-skill-memory
+description: Generate SKILL memory package from style reference for easy loading and consistent design system usage
+argument-hint: "[package-name] [--regenerate]"
+allowed-tools: Bash,Read,Write,TodoWrite
+auto-continue: true
+---
+
+# Memory: Style SKILL Memory Generator
+
+## Overview
+
+**Purpose**: Convert style reference package into SKILL memory for easy loading and context management.
+
+**Input**: Style reference package at `.workflow/reference_style/{package-name}/`
+
+**Output**: SKILL memory index at `.claude/skills/style-{package-name}/SKILL.md`
+
+**Use Case**: Load design system context when working with UI components, analyzing design patterns, or implementing style guidelines.
+
+**Key Features**:
+- Extracts primary design references (colors, typography, spacing, etc.)
+- Provides dynamic adjustment guidelines for design tokens
+- Includes prerequisites and tooling requirements (browsers, PostCSS, dark mode)
+- Progressive loading structure for efficient token usage
+- Complete implementation examples with React components
+- Interactive preview showcase
+
+---
+
+## Quick Reference
+
+### Command Syntax
+
+```bash
+/memory:style-skill-memory [package-name] [--regenerate]
+
+# Arguments
+package-name    Style reference package name (required)
+--regenerate    Force regenerate SKILL.md even if it exists (optional)
+```
+
+### Usage Examples
+
+```bash
+# Generate SKILL memory for package
+/memory:style-skill-memory main-app-style-v1
+
+# Regenerate SKILL memory
+/memory:style-skill-memory main-app-style-v1 --regenerate
+
+# Package name from current directory or default
+/memory:style-skill-memory
+```
+
+### Key Variables
+
+**Input Variables**:
+- `PACKAGE_NAME`: Style reference package name
+- `PACKAGE_DIR`: `.workflow/reference_style/${package_name}`
+- `SKILL_DIR`: `.claude/skills/style-${package_name}`
+- `REGENERATE`: `true` if --regenerate flag, `false` otherwise
+
+**Data Sources** (Phase 2):
+- `DESIGN_TOKENS_DATA`: Complete design-tokens.json content (from Read)
+- `LAYOUT_TEMPLATES_DATA`: Complete layout-templates.json content (from Read)
+- `ANIMATION_TOKENS_DATA`: Complete animation-tokens.json content (from Read, if exists)
+
+**Metadata** (Phase 2):
+- `COMPONENT_COUNT`: Total components
+- `UNIVERSAL_COUNT`: Universal components count
+- `SPECIALIZED_COUNT`: Specialized components count
+- `UNIVERSAL_COMPONENTS`: Universal component names (first 5)
+- `HAS_ANIMATIONS`: Whether animation-tokens.json exists
+
+**Analysis Output** (`DESIGN_ANALYSIS` - Phase 2):
+- `has_colors`: Colors exist
+- `color_semantic`: Has semantic naming (primary/secondary/accent)
+- `uses_oklch`: Uses modern color spaces (oklch, lab, etc.)
+- `has_dark_mode`: Has separate light/dark mode color tokens
+- `spacing_pattern`: Pattern type ("linear", "geometric", "custom")
+- `spacing_scale`: Actual scale values (e.g., [4, 8, 16, 32, 64])
+- `has_typography`: Typography system exists
+- `typography_hierarchy`: Has size scale for hierarchy
+- `uses_calc`: Uses calc() expressions in token values
+- `has_radius`: Border radius exists
+- `radius_style`: Style characteristic ("sharp" <4px, "moderate" 4-8px, "rounded" >8px)
+- `has_shadows`: Shadow system exists
+- `shadow_pattern`: Elevation naming pattern
+- `has_animations`: Animation tokens exist
+- `animation_range`: Duration range (fast to slow)
+- `easing_variety`: Types of easing functions
+
+### Common Errors
+
+| Error | Cause | Resolution |
+|-------|-------|------------|
+| Package not found | Invalid package name or doesn't exist | Run `/workflow:ui-design:codify-style` first |
+| SKILL already exists | SKILL.md already generated | Use `--regenerate` flag |
+| Missing layout-templates.json | Incomplete package | Verify package integrity, re-run codify-style |
+| Invalid JSON format | Corrupted package files | Regenerate package with codify-style |
+
+---
+
+## Execution Process
+
+### Phase 1: Validate Package
+
+**TodoWrite** (First Action):
+```json
+[
+  {
+    "content": "Validate package exists and check SKILL status",
+    "activeForm": "Validating package and SKILL status",
+    "status": "in_progress"
+  },
+  {
+    "content": "Read package data and analyze design system",
+    "activeForm": "Reading package data and analyzing design system",
+    "status": "pending"
+  },
+  {
+    "content": "Generate SKILL.md with design principles and token values",
+    "activeForm": "Generating SKILL.md with design principles and token values",
+    "status": "pending"
+  }
+]
+```
+
+**Step 1: Parse Package Name**
+
+```bash
+# Get package name from argument or auto-detect
+bash(echo "${package_name}" || basename "$(pwd)" | sed 's/^style-//')
+```
+
+**Step 2: Validate Package Exists**
+
+```bash
+bash(test -d .workflow/reference_style/${package_name} && echo "exists" || echo "missing")
+```
+
+**Error Handling**:
+```javascript
+if (package_not_exists) {
+  error("ERROR: Style reference package not found: ${package_name}")
+  error("HINT: Run '/workflow:ui-design:codify-style' first to create package")
+  error("Available packages:")
+  bash(ls -1 .workflow/reference_style/ 2>/dev/null || echo "  (none)")
+  exit(1)
+}
+```
+
+**Step 3: Check SKILL Already Exists**
+
+```bash
+bash(test -f .claude/skills/style-${package_name}/SKILL.md && echo "exists" || echo "missing")
+```
+
+**Decision Logic**:
+```javascript
+if (skill_exists && !regenerate_flag) {
+  echo("SKILL memory already exists for: ${package_name}")
+  echo("Use --regenerate to force regeneration")
+  exit(0)
+}
+
+if (regenerate_flag && skill_exists) {
+  echo("Regenerating SKILL memory for: ${package_name}")
+}
+```
+
+**TodoWrite Update**: Mark "Validate" as completed, "Read package data" as in_progress
+
+---
+
+### Phase 2: Read Package Data & Analyze Design System
+
+**Step 1: Read All JSON Files**
+
+```bash
+# Read layout templates
+Read(file_path=".workflow/reference_style/${package_name}/layout-templates.json")
+
+# Read design tokens
+Read(file_path=".workflow/reference_style/${package_name}/design-tokens.json")
+
+# Read animation tokens (if exists)
+bash(test -f .workflow/reference_style/${package_name}/animation-tokens.json && echo "exists" || echo "missing")
+Read(file_path=".workflow/reference_style/${package_name}/animation-tokens.json")  # if exists
+```
+
+**Step 2: Extract Metadata for Description**
+
+```bash
+# Count components and classify by type
+bash(jq '.layout_templates | length' layout-templates.json)
+bash(jq '[.layout_templates[] | select(.component_type == "universal")] | length' layout-templates.json)
+bash(jq '[.layout_templates[] | select(.component_type == "specialized")] | length' layout-templates.json)
+bash(jq -r '.layout_templates | to_entries[] | select(.value.component_type == "universal") | .key' layout-templates.json | head -5)
+```
+
+Store results in metadata variables (see [Key Variables](#key-variables))
+
+**Step 3: Analyze Design System for Dynamic Principles**
+
+Analyze design-tokens.json to extract characteristics and patterns:
+
+```bash
+# Color system characteristics
+bash(jq '.colors | keys' design-tokens.json)
+bash(jq '.colors | to_entries[0:2] | map(.value)' design-tokens.json)
+# Check for modern color spaces
+bash(jq '.colors | to_entries[] | .value | test("oklch|lab|lch")' design-tokens.json)
+# Check for dark mode variants
+bash(jq '.colors | keys | map(select(contains("dark") or contains("light")))' design-tokens.json)
+# → Store: has_colors, color_semantic, uses_oklch, has_dark_mode
+
+# Spacing pattern detection
+bash(jq '.spacing | to_entries | map(.value) | map(gsub("[^0-9.]"; "") | tonumber)' design-tokens.json)
+# Analyze pattern: linear (4-8-12-16) vs geometric (4-8-16-32) vs custom
+# → Store: spacing_pattern, spacing_scale
+
+# Typography characteristics
+bash(jq '.typography | keys | map(select(contains("family") or contains("weight")))' design-tokens.json)
+bash(jq '.typography | to_entries | map(select(.key | contains("size"))) | .[].value' design-tokens.json)
+# Check for calc() usage
+bash(jq '. | tostring | test("calc\\(")' design-tokens.json)
+# → Store: has_typography, typography_hierarchy, uses_calc
+
+# Border radius style
+bash(jq '.border_radius | to_entries | map(.value)' design-tokens.json)
+# Check range: small (sharp <4px) vs moderate (4-8px) vs large (rounded >8px)
+# → Store: has_radius, radius_style
+
+# Shadow characteristics
+bash(jq '.shadows | keys' design-tokens.json)
+bash(jq '.shadows | to_entries[0].value' design-tokens.json)
+# → Store: has_shadows, shadow_pattern
+
+# Animations (if available)
+bash(jq '.duration | to_entries | map(.value)' animation-tokens.json)
+bash(jq '.easing | keys' animation-tokens.json)
+# → Store: has_animations, animation_range, easing_variety
+```
+
+Store analysis results in `DESIGN_ANALYSIS` (see [Key Variables](#key-variables))
+
+**Note**: Analysis focuses on characteristics and patterns, not counts. Include technical feature detection (oklch, calc, dark mode) for Prerequisites section.
+
+**TodoWrite Update**: Mark "Read package data" as completed, "Generate SKILL.md" as in_progress
+
+---
+
+### Phase 3: Generate SKILL.md
+
+**Step 1: Create SKILL Directory**
+
+```bash
+bash(mkdir -p .claude/skills/style-${package_name})
+```
+
+**Step 2: Generate Intelligent Description**
+
+**Format**:
+```
+{package_name} project-independent design system with {universal_count} universal layout templates and interactive preview (located at .workflow/reference_style/{package_name}). Load when working with reusable UI components, design tokens, layout patterns, or implementing visual consistency. Excludes {specialized_count} project-specific components.
+```
+
+**Step 3: Load and Process SKILL.md Template**
+
+**⚠️ CRITICAL - Execute First**:
+```bash
+bash(cat ~/.claude/workflows/cli-templates/memory/style-skill-memory/skill-md-template.md)
+```
+
+**Template Processing**:
+1. **Replace variables**: Substitute all `{variable}` placeholders with actual values from Phase 2
+2. **Generate dynamic sections**:
+   - **Prerequisites & Tooling**: Generate based on `DESIGN_ANALYSIS` technical features (oklch, calc, dark mode)
+   - **Design Principles**: Generate based on `DESIGN_ANALYSIS` characteristics
+   - **Complete Implementation Example**: Include React component example with token adaptation
+   - **Design Token Values**: Iterate `DESIGN_TOKENS_DATA`, `ANIMATION_TOKENS_DATA` and display all key-value pairs with DEFAULT annotations
+3. **Write to file**: Use Write tool to save to `.claude/skills/style-{package_name}/SKILL.md`
+
+**Variable Replacement Map**:
+- `{package_name}` → PACKAGE_NAME
+- `{intelligent_description}` → Generated description from Step 2
+- `{component_count}` → COMPONENT_COUNT
+- `{universal_count}` → UNIVERSAL_COUNT
+- `{specialized_count}` → SPECIALIZED_COUNT
+- `{universal_components_list}` → UNIVERSAL_COMPONENTS (comma-separated)
+- `{has_animations}` → HAS_ANIMATIONS
+
+**Dynamic Content Generation**:
+
+See template file for complete structure. Key dynamic sections:
+
+1. **Prerequisites & Tooling** (based on DESIGN_ANALYSIS technical features):
+   - IF uses_oklch → Include PostCSS plugin requirement (`postcss-oklab-function`)
+   - IF uses_calc → Include preprocessor requirement for calc() expressions
+   - IF has_dark_mode → Include dark mode implementation mechanism (class or media query)
+   - ALWAYS include browser support, jq installation, and local server setup
+
+2. **Design Principles** (based on DESIGN_ANALYSIS):
+   - IF has_colors → Include "Color System" principle with semantic pattern
+   - IF spacing_pattern detected → Include "Spatial Rhythm" with unified scale description (actual token values)
+   - IF has_typography_hierarchy → Include "Typographic System" with scale examples
+   - IF has_radius → Include "Shape Language" with style characteristic
+   - IF has_shadows → Include "Depth & Elevation" with elevation pattern
+   - IF has_animations → Include "Motion & Timing" with duration range
+   - ALWAYS include "Accessibility First" principle
+
+3. **Design Token Values** (iterate from read data):
+   - Colors: Iterate `DESIGN_TOKENS_DATA.colors`
+   - Typography: Iterate `DESIGN_TOKENS_DATA.typography`
+   - Spacing: Iterate `DESIGN_TOKENS_DATA.spacing`
+   - Border Radius: Iterate `DESIGN_TOKENS_DATA.border_radius` with calc() explanations
+   - Shadows: Iterate `DESIGN_TOKENS_DATA.shadows` with DEFAULT token annotations
+   - Animations (if available): Iterate `ANIMATION_TOKENS_DATA.duration` and `ANIMATION_TOKENS_DATA.easing`
+
+**Step 4: Verify SKILL.md Created**
+
+```bash
+bash(test -f .claude/skills/style-${package_name}/SKILL.md && echo "success" || echo "failed")
+```
+
+**TodoWrite Update**: Mark all todos as completed
+
+---
+
+### Completion Message
+
+Display a simple completion message with key information:
+
+```
+✅ SKILL memory generated for style package: {package_name}
+
+📁 Location: .claude/skills/style-{package_name}/SKILL.md
+
+📊 Package Summary:
+   - {component_count} components ({universal_count} universal, {specialized_count} specialized)
+   - Design tokens: colors, typography, spacing, shadows{animations_note}
+
+💡 Usage: /memory:load-skill-memory style-{package_name} "your task description"
+```
+
+Variables: `{package_name}`, `{component_count}`, `{universal_count}`, `{specialized_count}`, `{animations_note}` (", animations" if exists)
+
+---
+
+## Implementation Details
+
+### Critical Rules
+
+1. **Check Before Generate**: Verify package exists before attempting SKILL generation
+2. **Respect Existing SKILL**: Don't overwrite unless --regenerate flag provided
+3. **Load Templates via cat**: Use `cat ~/.claude/workflows/cli-templates/memory/style-skill-memory/{template}` to load templates
+4. **Variable Substitution**: Replace all `{variable}` placeholders with actual values
+5. **Technical Feature Detection**: Analyze tokens for modern features (oklch, calc, dark mode) and generate appropriate Prerequisites section
+6. **Dynamic Content Generation**: Generate sections based on DESIGN_ANALYSIS characteristics
+7. **Unified Spacing Scale**: Use actual token values as primary scale reference, avoid contradictory pattern descriptions
+8. **Direct Iteration**: Iterate data structures (DESIGN_TOKENS_DATA, etc.) for token values
+9. **Annotate Special Tokens**: Add comments for DEFAULT tokens and calc() expressions
+10. **Embed jq Commands**: Include bash/jq commands in SKILL.md for dynamic loading
+11. **Progressive Loading**: Include all 3 levels (0-2) with specific jq commands
+12. **Complete Examples**: Include end-to-end implementation examples (React components)
+13. **Intelligent Description**: Extract component count and key features from metadata
+14. **Emphasize Flexibility**: Strongly warn against rigid copying - values are references for creative adaptation
+
+### Template Files Location
+
+
+```
+Phase 1: Validate
+  ├─ Parse package_name
+  ├─ Check PACKAGE_DIR exists
+  └─ Check SKILL_DIR exists (skip if exists and no --regenerate)
+
+Phase 2: Read & Analyze
+  ├─ Read design-tokens.json → DESIGN_TOKENS_DATA
+  ├─ Read layout-templates.json → LAYOUT_TEMPLATES_DATA
+  ├─ Read animation-tokens.json → ANIMATION_TOKENS_DATA (if exists)
+  ├─ Extract Metadata → COMPONENT_COUNT, UNIVERSAL_COUNT, etc.
+  └─ Analyze Design System → DESIGN_ANALYSIS (characteristics)
+
+Phase 3: Generate
+  ├─ Create SKILL directory
+  ├─ Generate intelligent description
+  ├─ Load SKILL.md template (cat command)
+  ├─ Replace variables and generate dynamic content
+  ├─ Write SKILL.md
+  ├─ Verify creation
+  ├─ Load completion message template (cat command)
+  └─ Display completion message
+```

.claude/commands/memory/tech-research-rules.md

@@ -0,0 +1,310 @@
+---
+name: tech-research-rules
+description: "3-phase orchestrator: extract tech stack → Exa research → generate path-conditional rules (auto-loaded by Claude Code)"
+argument-hint: "[session-id | tech-stack-name] [--regenerate] [--tool <gemini|qwen>]"
+allowed-tools: SlashCommand(*), TodoWrite(*), Bash(*), Read(*), Write(*), Task(*)
+---
+
+# Tech Stack Rules Generator
+
+## Overview
+
+**Purpose**: Generate multi-layered, path-conditional rules that Claude Code automatically loads based on file context.
+
+**Output Structure**:
+```
+.claude/rules/tech/{tech-stack}/
+├── core.md           # paths: **/*.{ext} - Core principles
+├── patterns.md       # paths: src/**/*.{ext} - Implementation patterns
+├── testing.md        # paths: **/*.{test,spec}.{ext} - Testing rules
+├── config.md         # paths: *.config.* - Configuration rules
+├── api.md            # paths: **/api/**/* - API rules (backend only)
+├── components.md     # paths: **/components/**/* - Component rules (frontend only)
+└── metadata.json     # Generation metadata
+```
+
+**Templates Location**: `~/.claude/workflows/cli-templates/prompts/rules/`
+
+---
+
+## Core Rules
+
+1. **Start Immediately**: First action is TodoWrite initialization
+2. **Path-Conditional Output**: Every rule file includes `paths` frontmatter
+3. **Template-Driven**: Agent reads templates before generating content
+4. **Agent Produces Files**: Agent writes all rule files directly
+5. **No Manual Loading**: Rules auto-activate when Claude works with matching files
+
+---
+
+## 3-Phase Execution
+
+### Phase 1: Prepare Context & Detect Tech Stack
+
+**Goal**: Detect input mode, extract tech stack info, determine file extensions
+
+**Input Mode Detection**:
+```bash
+input="$1"
+
+if [[ "$input" == WFS-* ]]; then
+  MODE="session"
+  SESSION_ID="$input"
+  # Read workflow-session.json to extract tech stack
+else
+  MODE="direct"
+  TECH_STACK_NAME="$input"
+fi
+```
+
+**Tech Stack Analysis**:
+```javascript
+// Decompose composite tech stacks
+// "typescript-react-nextjs" → ["typescript", "react", "nextjs"]
+
+const TECH_EXTENSIONS = {
+  "typescript": "{ts,tsx}",
+  "javascript": "{js,jsx}",
+  "python": "py",
+  "rust": "rs",
+  "go": "go",
+  "java": "java",
+  "csharp": "cs",
+  "ruby": "rb",
+  "php": "php"
+};
+
+const FRAMEWORK_TYPE = {
+  "react": "frontend",
+  "vue": "frontend",
+  "angular": "frontend",
+  "nextjs": "fullstack",
+  "nuxt": "fullstack",
+  "fastapi": "backend",
+  "express": "backend",
+  "django": "backend",
+  "rails": "backend"
+};
+```
+
+**Check Existing Rules**:
+```bash
+normalized_name=$(echo "$TECH_STACK_NAME" | tr '[:upper:]' '[:lower:]' | tr ' ' '-')
+rules_dir=".claude/rules/tech/${normalized_name}"
+existing_count=$(find "${rules_dir}" -name "*.md" 2>/dev/null | wc -l || echo 0)
+```
+
+**Skip Decision**:
+- If `existing_count > 0` AND no `--regenerate` → `SKIP_GENERATION = true`
+- If `--regenerate` → Delete existing and regenerate
+
+**Output Variables**:
+- `TECH_STACK_NAME`: Normalized name
+- `PRIMARY_LANG`: Primary language
+- `FILE_EXT`: File extension pattern
+- `FRAMEWORK_TYPE`: frontend | backend | fullstack | library
+- `COMPONENTS`: Array of tech components
+- `SKIP_GENERATION`: Boolean
+
+**TodoWrite**: Mark phase 1 completed
+
+---
+
+### Phase 2: Agent Produces Path-Conditional Rules
+
+**Skip Condition**: Skipped if `SKIP_GENERATION = true`
+
+**Goal**: Delegate to agent for Exa research and rule file generation
+
+**Template Files**:
+```
+~/.claude/workflows/cli-templates/prompts/rules/
+├── tech-rules-agent-prompt.txt  # Agent instructions
+├── rule-core.txt                # Core principles template
+├── rule-patterns.txt            # Implementation patterns template
+├── rule-testing.txt             # Testing rules template
+├── rule-config.txt              # Configuration rules template
+├── rule-api.txt                 # API rules template (backend)
+└── rule-components.txt          # Component rules template (frontend)
+```
+
+**Agent Task**:
+
+```javascript
+Task({
+  subagent_type: "general-purpose",
+  description: `Generate tech stack rules: ${TECH_STACK_NAME}`,
+  prompt: `
+You are generating path-conditional rules for Claude Code.
+
+## Context
+- Tech Stack: ${TECH_STACK_NAME}
+- Primary Language: ${PRIMARY_LANG}
+- File Extensions: ${FILE_EXT}
+- Framework Type: ${FRAMEWORK_TYPE}
+- Components: ${JSON.stringify(COMPONENTS)}
+- Output Directory: .claude/rules/tech/${TECH_STACK_NAME}/
+
+## Instructions
+
+Read the agent prompt template for detailed instructions:
+$(cat ~/.claude/workflows/cli-templates/prompts/rules/tech-rules-agent-prompt.txt)
+
+## Execution Steps
+
+1. Execute Exa research queries (see agent prompt)
+2. Read each rule template
+3. Generate rule files following template structure
+4. Write files to output directory
+5. Write metadata.json
+6. Report completion
+
+## Variable Substitutions
+
+Replace in templates:
+- {TECH_STACK_NAME} → ${TECH_STACK_NAME}
+- {PRIMARY_LANG} → ${PRIMARY_LANG}
+- {FILE_EXT} → ${FILE_EXT}
+- {FRAMEWORK_TYPE} → ${FRAMEWORK_TYPE}
+`
+})
+```
+
+**Completion Criteria**:
+- 4-6 rule files written with proper `paths` frontmatter
+- metadata.json written
+- Agent reports files created
+
+**TodoWrite**: Mark phase 2 completed
+
+---
+
+### Phase 3: Verify & Report
+
+**Goal**: Verify generated files and provide usage summary
+
+**Steps**:
+
+1. **Verify Files**:
+   ```bash
+   find ".claude/rules/tech/${TECH_STACK_NAME}" -name "*.md" -type f
+   ```
+
+2. **Validate Frontmatter**:
+   ```bash
+   head -5 ".claude/rules/tech/${TECH_STACK_NAME}/core.md"
+   ```
+
+3. **Read Metadata**:
+   ```javascript
+   Read(`.claude/rules/tech/${TECH_STACK_NAME}/metadata.json`)
+   ```
+
+4. **Generate Summary Report**:
+   ```
+   Tech Stack Rules Generated
+
+   Tech Stack: {TECH_STACK_NAME}
+   Location: .claude/rules/tech/{TECH_STACK_NAME}/
+
+   Files Created:
+   ├── core.md         → paths: **/*.{ext}
+   ├── patterns.md     → paths: src/**/*.{ext}
+   ├── testing.md      → paths: **/*.{test,spec}.{ext}
+   ├── config.md       → paths: *.config.*
+   ├── api.md          → paths: **/api/**/* (if backend)
+   └── components.md   → paths: **/components/**/* (if frontend)
+
+   Auto-Loading:
+   - Rules apply automatically when editing matching files
+   - No manual loading required
+
+   Example Activation:
+   - Edit src/components/Button.tsx → core.md + patterns.md + components.md
+   - Edit tests/api.test.ts → core.md + testing.md
+   - Edit package.json → config.md
+   ```
+
+**TodoWrite**: Mark phase 3 completed
+
+---
+
+## Path Pattern Reference
+
+| Pattern | Matches |
+|---------|---------|
+| `**/*.ts` | All .ts files |
+| `src/**/*` | All files under src/ |
+| `*.config.*` | Config files in root |
+| `**/*.{ts,tsx}` | .ts and .tsx files |
+
+| Tech Stack | Core Pattern | Test Pattern |
+|------------|--------------|--------------|
+| TypeScript | `**/*.{ts,tsx}` | `**/*.{test,spec}.{ts,tsx}` |
+| Python | `**/*.py` | `**/test_*.py, **/*_test.py` |
+| Rust | `**/*.rs` | `**/tests/**/*.rs` |
+| Go | `**/*.go` | `**/*_test.go` |
+
+---
+
+## Parameters
+
+```bash
+/memory:tech-research [session-id | "tech-stack-name"] [--regenerate]
+```
+
+**Arguments**:
+- **session-id**: `WFS-*` format - Extract from workflow session
+- **tech-stack-name**: Direct input - `"typescript"`, `"typescript-react"`
+- **--regenerate**: Force regenerate existing rules
+
+---
+
+## Examples
+
+### Single Language
+
+```bash
+/memory:tech-research "typescript"
+```
+
+**Output**: `.claude/rules/tech/typescript/` with 4 rule files
+
+### Frontend Stack
+
+```bash
+/memory:tech-research "typescript-react"
+```
+
+**Output**: `.claude/rules/tech/typescript-react/` with 5 rule files (includes components.md)
+
+### Backend Stack
+
+```bash
+/memory:tech-research "python-fastapi"
+```
+
+**Output**: `.claude/rules/tech/python-fastapi/` with 5 rule files (includes api.md)
+
+### From Session
+
+```bash
+/memory:tech-research WFS-user-auth-20251104
+```
+
+**Workflow**: Extract tech stack from session → Generate rules
+
+---
+
+## Comparison: Rules vs SKILL
+
+| Aspect | SKILL Memory | Rules |
+|--------|--------------|-------|
+| Loading | Manual: `Skill("tech")` | Automatic by path |
+| Scope | All files when loaded | Only matching files |
+| Granularity | Monolithic packages | Per-file-type |
+| Context | Full package | Only relevant rules |
+
+**When to Use**:
+- **Rules**: Tech stack conventions per file type
+- **SKILL**: Reference docs, APIs, examples for manual lookup

.claude/commands/memory/tech-research.md

@@ -1,477 +0,0 @@
----
-name: tech-research
-description: Generate tech stack SKILL packages using Exa research via agent delegation
-argument-hint: "[session-id | tech-stack-name] [--regenerate] [--tool <gemini|qwen>]"
-allowed-tools: SlashCommand(*), TodoWrite(*), Bash(*), Read(*), Write(*), Task(*)
----
-
-# Tech Stack Research SKILL Generator
-
-## Overview
-
-**Pure Orchestrator with Agent Delegation**: Prepares context paths and delegates ALL work to agent. Agent produces files directly.
-
-**Auto-Continue Workflow**: Runs fully autonomously once triggered. Each phase completes and automatically triggers the next phase.
-
-**Execution Paths**:
-- **Full Path**: All 3 phases (no existing SKILL OR `--regenerate` specified)
-- **Skip Path**: Phase 1 → Phase 3 (existing SKILL found AND no `--regenerate` flag)
-- **Phase 3 Always Executes**: SKILL index is always generated or updated
-
-**Agent Responsibility**:
-- Agent does ALL the work: context reading, Exa research, content synthesis, file writing
-- Orchestrator only provides context paths and waits for completion
-
-## Core Rules
-
-1. **Start Immediately**: First action is TodoWrite initialization, second action is Phase 1 execution
-2. **Context Path Delegation**: Pass session directory or tech stack name to agent, let agent do discovery
-3. **Agent Produces Files**: Agent directly writes all module files, orchestrator does NOT parse agent output
-4. **Auto-Continue**: After completing each phase, update TodoWrite and immediately execute next phase
-5. **No User Prompts**: Never ask user questions or wait for input between phases
-6. **Track Progress**: Update TodoWrite after EVERY phase completion before starting next phase
-7. **Lightweight Index**: Phase 3 only generates SKILL.md index by reading existing files
-
----
-
-## 3-Phase Execution
-
-### Phase 1: Prepare Context Paths
-
-**Goal**: Detect input mode, prepare context paths for agent, check existing SKILL
-
-**Input Mode Detection**:
-```bash
-# Get input parameter
-input="$1"
-
-# Detect mode
-if [[ "$input" == WFS-* ]]; then
-  MODE="session"
-  SESSION_ID="$input"
-  CONTEXT_PATH=".workflow/${SESSION_ID}"
-else
-  MODE="direct"
-  TECH_STACK_NAME="$input"
-  CONTEXT_PATH="$input"  # Pass tech stack name as context
-fi
-```
-
-**Check Existing SKILL**:
-```bash
-# For session mode, peek at session to get tech stack name
-if [[ "$MODE" == "session" ]]; then
-  bash(test -f ".workflow/${SESSION_ID}/workflow-session.json")
-  Read(.workflow/${SESSION_ID}/workflow-session.json)
-  # Extract tech_stack_name (minimal extraction)
-fi
-
-# Normalize and check
-normalized_name=$(echo "$TECH_STACK_NAME" | tr '[:upper:]' '[:lower:]' | tr ' ' '-')
-bash(test -d ".claude/skills/${normalized_name}" && echo "exists" || echo "not_exists")
-bash(find ".claude/skills/${normalized_name}" -name "*.md" 2>/dev/null | wc -l || echo 0)
-```
-
-**Skip Decision**:
-```javascript
-if (existing_files > 0 && !regenerate_flag) {
-  SKIP_GENERATION = true
-  message = "Tech stack SKILL already exists, skipping Phase 2. Use --regenerate to force regeneration."
-} else if (regenerate_flag) {
-  bash(rm -rf ".claude/skills/${normalized_name}")
-  SKIP_GENERATION = false
-  message = "Regenerating tech stack SKILL from scratch."
-} else {
-  SKIP_GENERATION = false
-  message = "No existing SKILL found, generating new tech stack documentation."
-}
-```
-
-**Output Variables**:
-- `MODE`: `session` or `direct`
-- `SESSION_ID`: Session ID (if session mode)
-- `CONTEXT_PATH`: Path to session directory OR tech stack name
-- `TECH_STACK_NAME`: Extracted or provided tech stack name
-- `SKIP_GENERATION`: Boolean - whether to skip Phase 2
-
-**TodoWrite**:
-- If skipping: Mark phase 1 completed, phase 2 completed, phase 3 in_progress
-- If not skipping: Mark phase 1 completed, phase 2 in_progress
-
----
-
-### Phase 2: Agent Produces All Files
-
-**Skip Condition**: Skipped if `SKIP_GENERATION = true`
-
-**Goal**: Delegate EVERYTHING to agent - context reading, Exa research, content synthesis, and file writing
-
-**Agent Task Specification**:
-
-```
-Task(
-  subagent_type: "general-purpose",
-  description: "Generate tech stack SKILL: {CONTEXT_PATH}",
-  prompt: "
-Generate a complete tech stack SKILL package with Exa research.
-
-**Context Provided**:
-- Mode: {MODE}
-- Context Path: {CONTEXT_PATH}
-
-**Templates Available**:
-- Module Format: ~/.claude/workflows/cli-templates/prompts/tech/tech-module-format.txt
-- SKILL Index: ~/.claude/workflows/cli-templates/prompts/tech/tech-skill-index.txt
-
-**Your Responsibilities**:
-
-1. **Extract Tech Stack Information**:
-
-   IF MODE == 'session':
-     - Read `.workflow/{SESSION_ID}/workflow-session.json`
-     - Read `.workflow/{SESSION_ID}/.process/context-package.json`
-     - Extract tech_stack: {language, frameworks, libraries}
-     - Build tech stack name: \"{language}-{framework1}-{framework2}\"
-     - Example: \"typescript-react-nextjs\"
-
-   IF MODE == 'direct':
-     - Tech stack name = CONTEXT_PATH
-     - Parse composite: split by '-' delimiter
-     - Example: \"typescript-react-nextjs\" → [\"typescript\", \"react\", \"nextjs\"]
-
-2. **Execute Exa Research** (4-6 parallel queries):
-
-   Base Queries (always execute):
-   - mcp__exa__get_code_context_exa(query: \"{tech} core principles best practices 2025\", tokensNum: 8000)
-   - mcp__exa__get_code_context_exa(query: \"{tech} common patterns architecture examples\", tokensNum: 7000)
-   - mcp__exa__web_search_exa(query: \"{tech} configuration setup tooling 2025\", numResults: 5)
-   - mcp__exa__get_code_context_exa(query: \"{tech} testing strategies\", tokensNum: 5000)
-
-   Component Queries (if composite):
-   - For each additional component:
-     mcp__exa__get_code_context_exa(query: \"{main_tech} {component} integration\", tokensNum: 5000)
-
-3. **Read Module Format Template**:
-
-   Read template for structure guidance:
-   ```bash
-   Read(~/.claude/workflows/cli-templates/prompts/tech/tech-module-format.txt)
-   ```
-
-4. **Synthesize Content into 6 Modules**:
-
-   Follow template structure from tech-module-format.txt:
-   - **principles.md** - Core concepts, philosophies (~3K tokens)
-   - **patterns.md** - Implementation patterns with code examples (~5K tokens)
-   - **practices.md** - Best practices, anti-patterns, pitfalls (~4K tokens)
-   - **testing.md** - Testing strategies, frameworks (~3K tokens)
-   - **config.md** - Setup, configuration, tooling (~3K tokens)
-   - **frameworks.md** - Framework integration (only if composite, ~4K tokens)
-
-   Each module follows template format:
-   - Frontmatter (YAML)
-   - Main sections with clear headings
-   - Code examples from Exa research
-   - Best practices sections
-   - References to Exa sources
-
-5. **Write Files Directly**:
-
-   ```javascript
-   // Create directory
-   bash(mkdir -p \".claude/skills/{tech_stack_name}\")
-
-   // Write each module file using Write tool
-   Write({ file_path: \".claude/skills/{tech_stack_name}/principles.md\", content: ... })
-   Write({ file_path: \".claude/skills/{tech_stack_name}/patterns.md\", content: ... })
-   Write({ file_path: \".claude/skills/{tech_stack_name}/practices.md\", content: ... })
-   Write({ file_path: \".claude/skills/{tech_stack_name}/testing.md\", content: ... })
-   Write({ file_path: \".claude/skills/{tech_stack_name}/config.md\", content: ... })
-   // Write frameworks.md only if composite
-
-   // Write metadata.json
-   Write({
-     file_path: \".claude/skills/{tech_stack_name}/metadata.json\",
-     content: JSON.stringify({
-       tech_stack_name,
-       components,
-       is_composite,
-       generated_at: timestamp,
-       source: \"exa-research\",
-       research_summary: { total_queries, total_sources }
-     })
-   })
-   ```
-
-6. **Report Completion**:
-
-   Provide summary:
-   - Tech stack name
-   - Files created (count)
-   - Exa queries executed
-   - Sources consulted
-
-**CRITICAL**:
-- MUST read external template files before generating content (step 3 for modules, step 4 for index)
-- You have FULL autonomy - read files, execute Exa, synthesize content, write files
-- Do NOT return JSON or structured data - produce actual .md files
-- Handle errors gracefully (Exa failures, missing files, template read failures)
-- If tech stack cannot be determined, ask orchestrator to clarify
-  "
-)
-```
-
-**Completion Criteria**:
-- Agent task executed successfully
-- 5-6 modular files written to `.claude/skills/{tech_stack_name}/`
-- metadata.json written
-- Agent reports completion
-
-**TodoWrite**: Mark phase 2 completed, phase 3 in_progress
-
----
-
-### Phase 3: Generate SKILL.md Index
-
-**Note**: This phase **ALWAYS executes** - generates or updates the SKILL index.
-
-**Goal**: Read generated module files and create SKILL.md index with loading recommendations
-
-**Steps**:
-
-1. **Verify Generated Files**:
-   ```bash
-   bash(find ".claude/skills/${TECH_STACK_NAME}" -name "*.md" -type f | sort)
-   ```
-
-2. **Read metadata.json**:
-   ```javascript
-   Read(.claude/skills/${TECH_STACK_NAME}/metadata.json)
-   // Extract: tech_stack_name, components, is_composite, research_summary
-   ```
-
-3. **Read Module Headers** (optional, first 20 lines):
-   ```javascript
-   Read(.claude/skills/${TECH_STACK_NAME}/principles.md, limit: 20)
-   // Repeat for other modules
-   ```
-
-4. **Read SKILL Index Template**:
-
-   ```javascript
-   Read(~/.claude/workflows/cli-templates/prompts/tech/tech-skill-index.txt)
-   ```
-
-5. **Generate SKILL.md Index**:
-
-   Follow template from tech-skill-index.txt with variable substitutions:
-   - `{TECH_STACK_NAME}`: From metadata.json
-   - `{MAIN_TECH}`: Primary technology
-   - `{ISO_TIMESTAMP}`: Current timestamp
-   - `{QUERY_COUNT}`: From research_summary
-   - `{SOURCE_COUNT}`: From research_summary
-   - Conditional sections for composite tech stacks
-
-   Template provides structure for:
-   - Frontmatter with metadata
-   - Overview and tech stack description
-   - Module organization (Core/Practical/Config sections)
-   - Loading recommendations (Quick/Implementation/Complete)
-   - Usage guidelines and auto-trigger keywords
-   - Research metadata and version history
-
-6. **Write SKILL.md**:
-   ```javascript
-   Write({
-     file_path: `.claude/skills/${TECH_STACK_NAME}/SKILL.md`,
-     content: generatedIndexMarkdown
-   })
-   ```
-
-**Completion Criteria**:
-- SKILL.md index written
-- All module files verified
-- Loading recommendations included
-
-**TodoWrite**: Mark phase 3 completed
-
-**Final Report**:
-```
-Tech Stack SKILL Package Complete
-
-Tech Stack: {TECH_STACK_NAME}
-Location: .claude/skills/{TECH_STACK_NAME}/
-
-Files: SKILL.md + 5-6 modules + metadata.json
-Exa Research: {queries} queries, {sources} sources
-
-Usage: Skill(command: "{TECH_STACK_NAME}")
-```
-
----
-
-## Implementation Details
-
-### TodoWrite Patterns
-
-**Initialization** (Before Phase 1):
-```javascript
-TodoWrite({todos: [
-  {"content": "Prepare context paths", "status": "in_progress", "activeForm": "Preparing context paths"},
-  {"content": "Agent produces all module files", "status": "pending", "activeForm": "Agent producing files"},
-  {"content": "Generate SKILL.md index", "status": "pending", "activeForm": "Generating SKILL index"}
-]})
-```
-
-**Full Path** (SKIP_GENERATION = false):
-```javascript
-// After Phase 1
-TodoWrite({todos: [
-  {"content": "Prepare context paths", "status": "completed", ...},
-  {"content": "Agent produces all module files", "status": "in_progress", ...},
-  {"content": "Generate SKILL.md index", "status": "pending", ...}
-]})
-
-// After Phase 2
-TodoWrite({todos: [
-  {"content": "Prepare context paths", "status": "completed", ...},
-  {"content": "Agent produces all module files", "status": "completed", ...},
-  {"content": "Generate SKILL.md index", "status": "in_progress", ...}
-]})
-
-// After Phase 3
-TodoWrite({todos: [
-  {"content": "Prepare context paths", "status": "completed", ...},
-  {"content": "Agent produces all module files", "status": "completed", ...},
-  {"content": "Generate SKILL.md index", "status": "completed", ...}
-]})
-```
-
-**Skip Path** (SKIP_GENERATION = true):
-```javascript
-// After Phase 1 (skip Phase 2)
-TodoWrite({todos: [
-  {"content": "Prepare context paths", "status": "completed", ...},
-  {"content": "Agent produces all module files", "status": "completed", ...},  // Skipped
-  {"content": "Generate SKILL.md index", "status": "in_progress", ...}
-]})
-```
-
-### Execution Flow
-
-**Full Path**:
-```
-User → TodoWrite Init → Phase 1 (prepare) → Phase 2 (agent writes files) → Phase 3 (write index) → Report
-```
-
-**Skip Path**:
-```
-User → TodoWrite Init → Phase 1 (detect existing) → Phase 3 (update index) → Report
-```
-
-### Error Handling
-
-**Phase 1 Errors**:
-- Invalid session ID: Report error, verify session exists
-- Missing context-package: Warn, fall back to direct mode
-- No tech stack detected: Ask user to specify tech stack name
-
-**Phase 2 Errors (Agent)**:
-- Agent task fails: Retry once, report if fails again
-- Exa API failures: Agent handles internally with retries
-- Incomplete results: Warn user, proceed with partial data if minimum sections available
-
-**Phase 3 Errors**:
-- Write failures: Report which files failed
-- Missing files: Note in SKILL.md, suggest regeneration
-
----
-
-## Parameters
-
-```bash
-/memory:tech-research [session-id | "tech-stack-name"] [--regenerate] [--tool <gemini|qwen>]
-```
-
-**Arguments**:
-- **session-id | tech-stack-name**: Input source (auto-detected by WFS- prefix)
-  - Session mode: `WFS-user-auth-v2` - Extract tech stack from workflow
-  - Direct mode: `"typescript"`, `"typescript-react-nextjs"` - User specifies
-- **--regenerate**: Force regenerate existing SKILL (deletes and recreates)
-- **--tool**: Reserved for future CLI integration (default: gemini)
-
----
-
-## Examples
-
-**Generated File Structure** (for all examples):
-```
-.claude/skills/{tech-stack}/
-├── SKILL.md           # Index (Phase 3)
-├── principles.md      # Agent (Phase 2)
-├── patterns.md        # Agent
-├── practices.md       # Agent
-├── testing.md         # Agent
-├── config.md          # Agent
-├── frameworks.md      # Agent (if composite)
-└── metadata.json      # Agent
-```
-
-### Direct Mode - Single Stack
-
-```bash
-/memory:tech-research "typescript"
-```
-
-**Workflow**:
-1. Phase 1: Detects direct mode, checks existing SKILL
-2. Phase 2: Agent executes 4 Exa queries, writes 5 modules
-3. Phase 3: Generates SKILL.md index
-
-### Direct Mode - Composite Stack
-
-```bash
-/memory:tech-research "typescript-react-nextjs"
-```
-
-**Workflow**:
-1. Phase 1: Decomposes into ["typescript", "react", "nextjs"]
-2. Phase 2: Agent executes 6 Exa queries (4 base + 2 components), writes 6 modules (adds frameworks.md)
-3. Phase 3: Generates SKILL.md index with framework integration
-
-### Session Mode - Extract from Workflow
-
-```bash
-/memory:tech-research WFS-user-auth-20251104
-```
-
-**Workflow**:
-1. Phase 1: Reads session, extracts tech stack: `python-fastapi-sqlalchemy`
-2. Phase 2: Agent researches Python + FastAPI + SQLAlchemy, writes 6 modules
-3. Phase 3: Generates SKILL.md index
-
-### Regenerate Existing
-
-```bash
-/memory:tech-research "react" --regenerate
-```
-
-**Workflow**:
-1. Phase 1: Deletes existing SKILL due to --regenerate
-2. Phase 2: Agent executes fresh Exa research (latest 2025 practices)
-3. Phase 3: Generates updated SKILL.md
-
-### Skip Path - Fast Update
-
-```bash
-/memory:tech-research "python"
-```
-
-**Scenario**: SKILL already exists with 7 files
-
-**Workflow**:
-1. Phase 1: Detects existing SKILL, sets SKIP_GENERATION = true
-2. Phase 2: **SKIPPED**
-3. Phase 3: Updates SKILL.md index only (5-10x faster)
-
-

.claude/commands/memory/update-full.md

@@ -1,6 +1,6 @@
 ---
 name: update-full
-description: Complete project-wide CLAUDE.md documentation update with agent-based parallel execution and tool fallback
+description: Update all CLAUDE.md files using layer-based execution (Layer 3→1) with batched agents (4 modules/agent) and gemini→qwen→codex fallback, <20 modules uses direct parallel
 argument-hint: "[--tool gemini|qwen|codex] [--path <directory>]"
 ---
 
@@ -36,13 +36,12 @@ Orchestrates project-wide CLAUDE.md updates using batched agent execution with a
 - **Use Case**: Deepest directories with unstructured file layouts
 - **Behavior**: Generates CLAUDE.md for current directory AND each subdirectory containing files
 - **Context**: All files in current directory tree (`@**/*`)
-- **Benefits**: Creates foundation documentation for upper layers to reference
+
 
 #### Single-Layer Strategy (Layers 1-2)
 - **Use Case**: Upper layers that aggregate from existing documentation
 - **Behavior**: Generates CLAUDE.md only for current directory
 - **Context**: Direct children CLAUDE.md files + current directory code files
-- **Benefits**: Minimal context consumption, clear layer separation
 
 ### Example Flow
 ```
@@ -95,14 +94,15 @@ src/ (depth 1) → SINGLE-LAYER STRATEGY
 
 ### Phase 1: Discovery & Analysis
 
-```bash
-# Cache git changes
-bash(git add -A 2>/dev/null || true)
+```javascript
+// Cache git changes
+Bash({command: "git add -A 2>/dev/null || true", run_in_background: false});
 
-# Get module structure
-bash(~/.claude/scripts/get_modules_by_depth.sh list)
-# OR with --path
-bash(cd <target-path> && ~/.claude/scripts/get_modules_by_depth.sh list)
+// Get module structure
+Bash({command: "ccw tool exec get_modules_by_depth '{\"format\":\"list\"}'", run_in_background: false});
+
+// OR with --path
+Bash({command: "cd <target-path> && ccw tool exec get_modules_by_depth '{\"format\":\"list\"}'", run_in_background: false});
 ```
 
 **Parse output** `depth:N|path:<PATH>|...` to extract module paths and count.
@@ -172,26 +172,23 @@ Update Plan:
 
 **Strategy**: Parallel execution within layer (max 4 concurrent), no agent overhead.
 
-```javascript
-// Group modules by LAYER (not depth)
-let modules_by_layer = group_by_layer(module_list);
-let tool_order = construct_tool_order(primary_tool);
+**CRITICAL**: All Bash commands use `run_in_background: false` for synchronous execution.
 
-// Process by LAYER (3 → 2 → 1), not by depth
+```javascript
 for (let layer of [3, 2, 1]) {
   if (modules_by_layer[layer].length === 0) continue;
-
   let batches = batch_modules(modules_by_layer[layer], 4);
 
   for (let batch of batches) {
     let parallel_tasks = batch.map(module => {
       return async () => {
-        // Auto-determine strategy based on depth
         let strategy = module.depth >= 3 ? "multi-layer" : "single-layer";
-
         for (let tool of tool_order) {
-          let exit_code = bash(`cd ${module.path} && ~/.claude/scripts/update_module_claude.sh "${strategy}" "." "${tool}"`);
-          if (exit_code === 0) {
+          Bash({
+            command: `cd ${module.path} && ccw tool exec update_module_claude '{"strategy":"${strategy}","path":".","tool":"${tool}"}'`,
+            run_in_background: false
+          });
+          if (bash_result.exit_code === 0) {
             report(`✅ ${module.path} (Layer ${layer}) updated with ${tool}`);
             return true;
           }
@@ -200,7 +197,6 @@ for (let layer of [3, 2, 1]) {
         return false;
       };
     });
-
     await Promise.all(parallel_tasks.map(task => task()));
   }
 }
@@ -248,14 +244,17 @@ MODULES:
 
 TOOLS (try in order): {{tool_1}}, {{tool_2}}, {{tool_3}}
 
-EXECUTION SCRIPT: ~/.claude/scripts/update_module_claude.sh
+EXECUTION SCRIPT: ccw tool exec update_module_claude
   - Accepts strategy parameter: multi-layer | single-layer
   - Tool execution via direct CLI commands (gemini/qwen/codex)
 
 EXECUTION FLOW (for each module):
   1. Tool fallback loop (exit on first success):
      for tool in {{tool_1}} {{tool_2}} {{tool_3}}; do
-       bash(cd "{{module_path}}" && ~/.claude/scripts/update_module_claude.sh "{{strategy}}" "." "${tool}")
+       Bash({
+         command: `cd "{{module_path}}" && ccw tool exec update_module_claude '{"strategy":"{{strategy}}","path":".","tool":"${tool}"}'`,
+         run_in_background: false
+       })
        exit_code=$?
 
        if [ $exit_code -eq 0 ]; then
@@ -287,12 +286,12 @@ REPORTING FORMAT:
 ```
 ### Phase 4: Safety Verification
 
-```bash
-# Check only CLAUDE.md modified
-bash(git diff --cached --name-only | grep -v "CLAUDE.md" || echo "Only CLAUDE.md files modified")
+```javascript
+// Check only CLAUDE.md files modified
+Bash({command: 'git diff --cached --name-only | grep -v "CLAUDE.md" || echo "Only CLAUDE.md files modified"', run_in_background: false});
 
-# Display status
-bash(git status --short)
+// Display status
+Bash({command: "git status --short", run_in_background: false});
 ```
 
 **Result Summary**:

.claude/commands/memory/update-related.md

@@ -1,6 +1,6 @@
 ---
 name: update-related
-description: Context-aware CLAUDE.md documentation updates based on recent changes with agent-based execution and tool fallback
+description: Update CLAUDE.md for git-changed modules using batched agent execution (4 modules/agent) with gemini→qwen→codex fallback, <15 modules uses direct execution
 argument-hint: "[--tool gemini|qwen|codex]"
 ---
 
@@ -39,12 +39,12 @@ Orchestrates context-aware CLAUDE.md updates for changed modules using batched a
 
 ## Phase 1: Change Detection & Analysis
 
-```bash
-# Detect changed modules (no index refresh needed)
-bash(~/.claude/scripts/detect_changed_modules.sh list)
+```javascript
+// Detect changed modules
+Bash({command: "ccw tool exec detect_changed_modules '{\"format\":\"list\"}'", run_in_background: false});
 
-# Cache git changes
-bash(git add -A 2>/dev/null || true)
+// Cache git changes
+Bash({command: "git add -A 2>/dev/null || true", run_in_background: false});
 ```
 
 **Parse output** `depth:N|path:<PATH>|change:<TYPE>` to extract affected modules.
@@ -89,47 +89,36 @@ Related Update Plan:
 
 ## Phase 3A: Direct Execution (<15 modules)
 
-**Strategy**: Parallel execution within depth (max 4 concurrent), no agent overhead, tool fallback per module.
+**Strategy**: Parallel execution within depth (max 4 concurrent), no agent overhead.
+
+**CRITICAL**: All Bash commands use `run_in_background: false` for synchronous execution.
 
 ```javascript
-let modules_by_depth = group_by_depth(changed_modules);
-let tool_order = construct_tool_order(primary_tool);
-
 for (let depth of sorted_depths.reverse()) {  // N → 0
-  let modules = modules_by_depth[depth];
-  let batches = batch_modules(modules, 4);  // Split into groups of 4
+  let batches = batch_modules(modules_by_depth[depth], 4);
 
   for (let batch of batches) {
-    // Execute batch in parallel (max 4 concurrent)
     let parallel_tasks = batch.map(module => {
       return async () => {
-        let success = false;
         for (let tool of tool_order) {
-          let exit_code = bash(cd ${module.path} && ~/.claude/scripts/update_module_claude.sh "single-layer" "." "${tool}");
-          if (exit_code === 0) {
-            report("${module.path} updated with ${tool}");
-            success = true;
-            break;
+          Bash({
+            command: `cd ${module.path} && ccw tool exec update_module_claude '{"strategy":"single-layer","path":".","tool":"${tool}"}'`,
+            run_in_background: false
+          });
+          if (bash_result.exit_code === 0) {
+            report(`✅ ${module.path} updated with ${tool}`);
+            return true;
           }
         }
-        if (!success) {
-          report("FAILED: ${module.path} failed all tools");
-        }
+        report(`❌ FAILED: ${module.path} failed all tools`);
+        return false;
       };
     });
-
-    await Promise.all(parallel_tasks.map(task => task()));  // Run batch in parallel
+    await Promise.all(parallel_tasks.map(task => task()));
   }
 }
 ```
 
-**Benefits**:
-- No agent startup overhead
-- Parallel execution within depth (max 4 concurrent)
-- Tool fallback still applies per module
-- Faster for small changesets (<15 modules)
-- Same batching strategy as Phase 3B but without agent layer
-
 ---
 
 ## Phase 3B: Agent Batch Execution (≥15 modules)
@@ -193,19 +182,27 @@ TOOLS (try in order):
 
 EXECUTION:
 For each module above:
-  1. cd "{{module_path}}"
-  2. Try tool 1:
-     bash(cd "{{module_path}}" && ~/.claude/scripts/update_module_claude.sh "single-layer" "." "{{tool_1}}")
-     → Success: Report "{{module_path}} updated with {{tool_1}}", proceed to next module
+  1. Try tool 1:
+     Bash({
+       command: `cd "{{module_path}}" && ccw tool exec update_module_claude '{"strategy":"single-layer","path":".","tool":"{{tool_1}}"}'`,
+       run_in_background: false
+     })
+     → Success: Report "✅ {{module_path}} updated with {{tool_1}}", proceed to next module
      → Failure: Try tool 2
-  3. Try tool 2:
-     bash(cd "{{module_path}}" && ~/.claude/scripts/update_module_claude.sh "single-layer" "." "{{tool_2}}")
-     → Success: Report "{{module_path}} updated with {{tool_2}}", proceed to next module
+  2. Try tool 2:
+     Bash({
+       command: `cd "{{module_path}}" && ccw tool exec update_module_claude '{"strategy":"single-layer","path":".","tool":"{{tool_2}}"}'`,
+       run_in_background: false
+     })
+     → Success: Report "✅ {{module_path}} updated with {{tool_2}}", proceed to next module
      → Failure: Try tool 3
-  4. Try tool 3:
-     bash(cd "{{module_path}}" && ~/.claude/scripts/update_module_claude.sh "single-layer" "." "{{tool_3}}")
-     → Success: Report "{{module_path}} updated with {{tool_3}}", proceed to next module
-     → Failure: Report "FAILED: {{module_path}} failed all tools", proceed to next module
+  3. Try tool 3:
+     Bash({
+       command: `cd "{{module_path}}" && ccw tool exec update_module_claude '{"strategy":"single-layer","path":".","tool":"{{tool_3}}"}'`,
+       run_in_background: false
+     })
+     → Success: Report "✅ {{module_path}} updated with {{tool_3}}", proceed to next module
+     → Failure: Report "❌ FAILED: {{module_path}} failed all tools", proceed to next module
 
 REPORTING:
 Report final summary with:
@@ -213,30 +210,16 @@ Report final summary with:
 - Successful: Y modules
 - Failed: Z modules
 - Tool usage: {{tool_1}}:X, {{tool_2}}:Y, {{tool_3}}:Z
-- Detailed results for each module
 ```
 
-### Example Execution
-
-**Depth 3 (new module)**:
-```javascript
-Task(subagent_type="memory-bridge", batch=[./src/api/auth], mode="related")
-```
-
-**Benefits**:
-- 4 modules → 1 agent (75% reduction)
-- Parallel batches, sequential within batch
-- Each module gets full fallback chain
-- Context-aware updates based on git changes
-
 ## Phase 4: Safety Verification
 
-```bash
-# Check only CLAUDE.md modified
-bash(git diff --cached --name-only | grep -v "CLAUDE.md" || echo "Only CLAUDE.md files modified")
+```javascript
+// Check only CLAUDE.md modified
+Bash({command: 'git diff --cached --name-only | grep -v "CLAUDE.md" || echo "Only CLAUDE.md files modified"', run_in_background: false});
 
-# Display statistics
-bash(git diff --stat)
+// Display statistics
+Bash({command: "git diff --stat", run_in_background: false});
 ```
 
 **Aggregate results**:

.claude/commands/memory/workflow-skill-memory.md

@@ -1,6 +1,6 @@
 ---
 name: workflow-skill-memory
-description: Generate SKILL package from archived workflow sessions for progressive context loading
+description: Process WFS-* archived sessions using universal-executor agents with Gemini analysis to generate workflow-progress SKILL package (sessions-timeline, lessons, conflicts)
 argument-hint: "session <session-id> | all"
 allowed-tools: Task(*), TodoWrite(*), Bash(*), Read(*), Write(*)
 ---
@@ -187,7 +187,7 @@ Objectives:
 
 3. Use Gemini for aggregation (optional):
    Command pattern:
-   cd .workflow/.archives/{session_id} && gemini -p "
+   ccw cli -p "
    PURPOSE: Extract lessons and conflicts from workflow session
    TASK:
    • Analyze IMPL_PLAN and lessons from manifest
@@ -198,7 +198,7 @@ Objectives:
    CONTEXT: @IMPL_PLAN.md @workflow-session.json
    EXPECTED: Structured lessons and conflicts in JSON format
    RULES: Template reference from skill-aggregation.txt
-   "
+   " --tool gemini --mode analysis --cd .workflow/.archives/{session_id}
 
 3.5. **Generate SKILL.md Description** (CRITICAL for auto-loading):
 
@@ -334,7 +334,7 @@ Objectives:
    - Sort sessions by date
 
 2. Use Gemini for final aggregation:
-   gemini -p "
+   ccw cli -p "
    PURPOSE: Aggregate lessons and conflicts from all workflow sessions
    TASK:
    • Group successes by functional domain
@@ -345,7 +345,7 @@ Objectives:
    CONTEXT: [Provide aggregated JSON data]
    EXPECTED: Final aggregated structure for SKILL documents
    RULES: Template reference from skill-aggregation.txt
-   "
+   " --tool gemini --mode analysis
 
 3. Read templates for formatting (same 4 templates as single mode)
 

.claude/commands/task/breakdown.md

@@ -1,6 +1,6 @@
 ---
 name: breakdown
-description: Intelligent task decomposition with context-aware subtask generation
+description: Decompose complex task into subtasks with dependency mapping, creates child task JSONs with parent references and execution order
 argument-hint: "task-id"
 ---
 

.claude/commands/task/create.md

@@ -1,6 +1,6 @@
 ---
 name: create
-description: Create implementation tasks with automatic context awareness
+description: Generate task JSON from natural language description with automatic file pattern detection, scope inference, and dependency analysis
 argument-hint: "\"task title\""
 ---
 

.claude/commands/task/execute.md

@@ -1,6 +1,6 @@
 ---
 name: execute
-description: Execute tasks with appropriate agents and context-aware orchestration
+description: Execute task JSON using appropriate agent (@doc-generator/@implementation-agent/@test-agent) with pre-analysis context loading and status tracking
 argument-hint: "task-id"
 ---
 
@@ -199,10 +199,10 @@ This is the simplified data structure loaded to provide context for task executi
     "session": "WFS-user-auth",
     "phase": "IMPLEMENT",
     "session_context": {
-      "workflow_directory": ".workflow/WFS-user-auth/",
-      "todo_list_location": ".workflow/WFS-user-auth/TODO_LIST.md",
-      "summaries_directory": ".workflow/WFS-user-auth/.summaries/",
-      "task_json_location": ".workflow/WFS-user-auth/.task/"
+      "workflow_directory": ".workflow/active/WFS-user-auth/",
+      "todo_list_location": ".workflow/active/WFS-user-auth/TODO_LIST.md",
+      "summaries_directory": ".workflow/active/WFS-user-auth/.summaries/",
+      "task_json_location": ".workflow/active/WFS-user-auth/.task/"
     }
   },
   "execution": {

.claude/commands/task/replan.md

@@ -1,12 +1,20 @@
 ---
 name: replan
-description: Replan individual tasks with detailed user input and change tracking
+description: Update task JSON with new requirements or batch-update multiple tasks from verification report, tracks changes in task-changes.json
 argument-hint: "task-id [\"text\"|file.md] | --batch [verification-report.md]"
 allowed-tools: Read(*), Write(*), Edit(*), TodoWrite(*), Glob(*), Bash(*)
 ---
 
 # Task Replan Command (/task:replan)
 
+> **⚠️ DEPRECATION NOTICE**: This command is maintained for backward compatibility. For new workflows, use `/workflow:replan` which provides:
+> - Session-level replanning with comprehensive artifact updates
+> - Interactive boundary clarification
+> - Updates to IMPL_PLAN.md, TODO_LIST.md, and session metadata
+> - Better integration with workflow sessions
+>
+> **Migration**: Replace `/task:replan IMPL-1 "changes"` with `/workflow:replan IMPL-1 "changes"`
+
 ## Overview
 Replans individual tasks or batch processes multiple tasks with change tracking and backup management.
 
@@ -14,9 +22,6 @@ Replans individual tasks or batch processes multiple tasks with change tracking
 - **Single Task Mode**: Replan one task with specific changes
 - **Batch Mode**: Process multiple tasks from action-plan verification report
 
-## Core Principles
-**Task System:** @~/.claude/workflows/task-core.md
-
 ## Key Features
 - **Single/Batch Operations**: Single task or multiple tasks from verification report
 - **Multiple Input Sources**: Text, files, or verification report
@@ -274,7 +279,7 @@ Backup saved to .task/backup/IMPL-2-v1.0.json
 
 ### Batch Mode - From Verification Report
 ```bash
-/task:replan --batch .workflow/WFS-{session}/.process/ACTION_PLAN_VERIFICATION.md
+/task:replan --batch .workflow/active/WFS-{session}/.process/ACTION_PLAN_VERIFICATION.md
 
 Parsing verification report...
 Found 4 tasks requiring replanning:

.claude/commands/version.md

@@ -1,6 +1,6 @@
 ---
 name: version
-description: Display version information and check for updates
+description: Display Claude Code version information and check for updates
 allowed-tools: Bash(*)
 ---
 

.claude/commands/workflow/action-plan-verify.md

@@ -1,6 +1,6 @@
 ---
 name: action-plan-verify
-description: Perform non-destructive cross-artifact consistency and quality analysis of IMPL_PLAN.md and task.json before execution
+description: Perform non-destructive cross-artifact consistency analysis between IMPL_PLAN.md and task JSONs with quality gate validation
 argument-hint: "[optional: --session session-id]"
 allowed-tools: Read(*), TodoWrite(*), Glob(*), Bash(*)
 ---
@@ -32,26 +32,32 @@ Identify inconsistencies, duplications, ambiguities, and underspecified items be
 IF --session parameter provided:
     session_id = provided session
 ELSE:
-    CHECK: .workflow/.active-* marker files
-    IF active_session EXISTS:
-        session_id = get_active_session()
-    ELSE:
+    # Auto-detect active session
+    active_sessions = bash(find .workflow/active/ -name "WFS-*" -type d 2>/dev/null)
+    IF active_sessions is empty:
         ERROR: "No active workflow session found. Use --session <session-id>"
         EXIT
+    ELSE IF active_sessions has multiple entries:
+        # Use most recently modified session
+        session_id = bash(ls -td .workflow/active/WFS-*/ 2>/dev/null | head -1 | xargs basename)
+    ELSE:
+        session_id = basename(active_sessions[0])
 
 # Derive absolute paths
-session_dir = .workflow/WFS-{session}
+session_dir = .workflow/active/WFS-{session}
 brainstorm_dir = session_dir/.brainstorming
 task_dir = session_dir/.task
 
 # Validate required artifacts
-SYNTHESIS = brainstorm_dir/role analysis documents
+# Note: "role analysis documents" refers to [role]/analysis.md files (e.g., product-manager/analysis.md)
+SYNTHESIS_DIR = brainstorm_dir  # Contains role analysis files: */analysis.md
 IMPL_PLAN = session_dir/IMPL_PLAN.md
 TASK_FILES = Glob(task_dir/*.json)
 
 # Abort if missing
-IF NOT EXISTS(SYNTHESIS):
-    ERROR: "role analysis documents not found. Run /workflow:brainstorm:synthesis first"
+SYNTHESIS_FILES = Glob(brainstorm_dir/*/analysis.md)
+IF SYNTHESIS_FILES.count == 0:
+    ERROR: "No role analysis documents found in .brainstorming/*/analysis.md. Run /workflow:brainstorm:synthesis first"
     EXIT
 
 IF NOT EXISTS(IMPL_PLAN):
@@ -95,7 +101,7 @@ Load only minimal necessary context from each artifact:
 - Dependencies (depends_on, blocks)
 - Context (requirements, focus_paths, acceptance, artifacts)
 - Flow control (pre_analysis, implementation_approach)
-- Meta (complexity, priority, use_codex)
+- Meta (complexity, priority)
 
 ### 3. Build Semantic Models
 
@@ -135,27 +141,27 @@ Focus on high-signal findings. Limit to 50 findings total; aggregate remainder i
 - **Unmapped Tasks**: Tasks with no clear requirement linkage
 - **NFR Coverage Gaps**: Non-functional requirements (performance, security, scalability) not reflected in tasks
 
-#### B. Consistency Validation
+#### C. Consistency Validation
 
 - **Requirement Conflicts**: Tasks contradicting synthesis requirements
 - **Architecture Drift**: IMPL_PLAN architecture not matching synthesis ADRs
 - **Terminology Drift**: Same concept named differently across IMPL_PLAN and tasks
 - **Data Model Inconsistency**: Tasks referencing entities/fields not in synthesis data model
 
-#### C. Dependency Integrity
+#### D. Dependency Integrity
 
 - **Circular Dependencies**: Task A depends on B, B depends on C, C depends on A
 - **Missing Dependencies**: Task requires outputs from another task but no explicit dependency
 - **Broken Dependencies**: Task depends on non-existent task ID
 - **Logical Ordering Issues**: Implementation tasks before foundational setup without dependency note
 
-#### D. Synthesis Alignment
+#### E. Synthesis Alignment
 
 - **Priority Conflicts**: High-priority synthesis requirements mapped to low-priority tasks
 - **Success Criteria Mismatch**: IMPL_PLAN success criteria not covering synthesis acceptance criteria
 - **Risk Mitigation Gaps**: Critical risks in synthesis without corresponding mitigation tasks
 
-#### E. Task Specification Quality
+#### F. Task Specification Quality
 
 - **Ambiguous Focus Paths**: Tasks with vague or missing focus_paths
 - **Underspecified Acceptance**: Tasks without clear acceptance criteria
@@ -163,12 +169,12 @@ Focus on high-signal findings. Limit to 50 findings total; aggregate remainder i
 - **Weak Flow Control**: Tasks without clear implementation_approach or pre_analysis steps
 - **Missing Target Files**: Tasks without flow_control.target_files specification
 
-#### F. Duplication Detection
+#### G. Duplication Detection
 
 - **Overlapping Task Scope**: Multiple tasks with nearly identical descriptions
 - **Redundant Requirements Coverage**: Same requirement covered by multiple tasks without clear partitioning
 
-#### G. Feasibility Assessment
+#### H. Feasibility Assessment
 
 - **Complexity Misalignment**: Task marked "simple" but requires multiple file modifications
 - **Resource Conflicts**: Parallel tasks requiring same resources/files
@@ -203,7 +209,9 @@ Use this heuristic to prioritize findings:
 
 ### 6. Produce Compact Analysis Report
 
-Output a Markdown report (no file writes) with the following structure:
+**Report Generation**: Generate report content and save to file.
+
+Output a Markdown report with the following structure:
 
 ```markdown
 ## Action Plan Verification Report
@@ -217,7 +225,11 @@ Output a Markdown report (no file writes) with the following structure:
 ### Executive Summary
 
 - **Overall Risk Level**: CRITICAL | HIGH | MEDIUM | LOW
-- **Recommendation**: BLOCK_EXECUTION | PROCEED_WITH_FIXES | PROCEED_WITH_CAUTION | PROCEED
+- **Recommendation**: (See decision matrix below)
+  - BLOCK_EXECUTION: Critical issues exist (must fix before proceeding)
+  - PROCEED_WITH_FIXES: High issues exist, no critical (fix recommended before execution)
+  - PROCEED_WITH_CAUTION: Medium issues only (proceed with awareness)
+  - PROCEED: Low issues only or no issues (safe to execute)
 - **Critical Issues**: {count}
 - **High Issues**: {count}
 - **Medium Issues**: {count}
@@ -322,18 +334,31 @@ Output a Markdown report (no file writes) with the following structure:
 
 #### Action Recommendations
 
-**If CRITICAL Issues Exist**:
-- **BLOCK EXECUTION** - Resolve critical issues before proceeding
-- Use TodoWrite to track all required fixes
-- Fix broken dependencies and circular references
+**Recommendation Decision Matrix**:
 
-**If Only HIGH/MEDIUM/LOW Issues**:
-- **PROCEED WITH CAUTION** - Fix high-priority issues first
-- Use TodoWrite to systematically track and complete all improvements
+| Condition | Recommendation | Action |
+|-----------|----------------|--------|
+| Critical > 0 | BLOCK_EXECUTION | Must resolve all critical issues before proceeding |
+| Critical = 0, High > 0 | PROCEED_WITH_FIXES | Fix high-priority issues before execution |
+| Critical = 0, High = 0, Medium > 0 | PROCEED_WITH_CAUTION | Proceed with awareness of medium issues |
+| Only Low or None | PROCEED | Safe to execute workflow |
+
+**If CRITICAL Issues Exist** (BLOCK_EXECUTION):
+- Resolve all critical issues before proceeding
+- Use TodoWrite to track required fixes
+- Fix broken dependencies and circular references first
+
+**If HIGH Issues Exist** (PROCEED_WITH_FIXES):
+- Fix high-priority issues before execution
+- Use TodoWrite to systematically track and complete improvements
+
+**If Only MEDIUM/LOW Issues** (PROCEED_WITH_CAUTION / PROCEED):
+- Can proceed with execution
+- Address issues during or after implementation
 
 #### TodoWrite-Based Remediation Workflow
 
-**Report Location**: `.workflow/WFS-{session}/.process/ACTION_PLAN_VERIFICATION.md`
+**Report Location**: `.workflow/active/WFS-{session}/.process/ACTION_PLAN_VERIFICATION.md`
 
 **Recommended Workflow**:
 1. **Create TodoWrite Task List**: Extract all findings from report
@@ -359,13 +384,18 @@ Priority Order:
 
 ### 7. Save Report and Execute TodoWrite-Based Remediation
 
-**Save Analysis Report**:
+**Step 7.1: Save Analysis Report**:
 ```bash
-report_path = ".workflow/WFS-{session}/.process/ACTION_PLAN_VERIFICATION.md"
+report_path = ".workflow/active/WFS-{session}/.process/ACTION_PLAN_VERIFICATION.md"
 Write(report_path, full_report_content)
 ```
 
-**After Report Generation**:
+**Step 7.2: Display Report Summary to User**:
+- Show executive summary with counts
+- Display recommendation (BLOCK/PROCEED_WITH_FIXES/PROCEED_WITH_CAUTION/PROCEED)
+- List critical and high issues if any
+
+**Step 7.3: After Report Generation**:
 
 1. **Extract Findings**: Parse all issues by severity
 2. **Create TodoWrite Task List**: Convert findings to actionable todos
@@ -404,12 +434,12 @@ TodoWrite([
 **File Modification Workflow**:
 ```bash
 # For task JSON modifications:
-1. Read(.workflow/WFS-{session}/.task/IMPL-X.Y.json)
+1. Read(.workflow/active/WFS-{session}/.task/IMPL-X.Y.json)
 2. Edit() to apply fixes
 3. Mark todo as completed
 
 # For IMPL_PLAN modifications:
-1. Read(.workflow/WFS-{session}/IMPL_PLAN.md)
+1. Read(.workflow/active/WFS-{session}/IMPL_PLAN.md)
 2. Edit() to apply strategic changes
 3. Mark todo as completed
 ```

.claude/commands/workflow/brainstorm/api-designer.md

@@ -1,6 +1,6 @@
 ---
 name: api-designer
-description: Generate or update api-designer/analysis.md addressing guidance-specification discussion points
+description: Generate or update api-designer/analysis.md addressing guidance-specification discussion points for API design perspective
 argument-hint: "optional topic - uses existing framework if available"
 allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ---
@@ -46,10 +46,10 @@ allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ### Phase 1: Session & Framework Detection
 ```bash
 # Check active session and framework
-CHECK: .workflow/.active-* marker files
+CHECK: find .workflow/active/ -name "WFS-*" -type d
 IF active_session EXISTS:
     session_id = get_active_session()
-    brainstorm_dir = .workflow/WFS-{session}/.brainstorming/
+    brainstorm_dir = .workflow/active/WFS-{session}/.brainstorming/
 
     CHECK: brainstorm_dir/guidance-specification.md
     IF EXISTS:
@@ -81,6 +81,7 @@ ELSE:
 **Framework-Based Analysis** (when guidance-specification.md exists):
 ```bash
 Task(subagent_type="conceptual-planning-agent",
+     run_in_background=false,
      prompt="Generate API designer analysis addressing topic framework
 
      ## Framework Integration Required
@@ -136,6 +137,7 @@ Task(subagent_type="conceptual-planning-agent",
 # For existing analysis updates
 IF update_mode = "incremental":
     Task(subagent_type="conceptual-planning-agent",
+         run_in_background=false,
          prompt="Update existing API designer analysis
 
          ## Current Analysis Context
@@ -162,7 +164,7 @@ IF update_mode = "incremental":
 
 ### Output Files
 ```
-.workflow/WFS-[topic]/.brainstorming/
+.workflow/active/WFS-[topic]/.brainstorming/
 ├── guidance-specification.md          # Input: Framework (if exists)
 └── api-designer/
     └── analysis.md            # ★ OUTPUT: Framework-based analysis
@@ -181,7 +183,7 @@ IF update_mode = "incremental":
 Session detection and selection:
 ```bash
 # Check for active sessions
-active_sessions=$(find .workflow -name ".active-*" 2>/dev/null)
+active_sessions=$(find .workflow/active/ -name "WFS-*" -type d 2>/dev/null)
 if [ multiple_sessions ]; then
   prompt_user_to_select_session()
 else
@@ -280,7 +282,7 @@ TodoWrite tracking for two-step process:
 
 ### Output Location
 ```
-.workflow/WFS-{topic-slug}/.brainstorming/api-designer/
+.workflow/active/WFS-{topic-slug}/.brainstorming/api-designer/
 ├── analysis.md                 # Primary API design analysis
 ├── api-specification.md        # Detailed endpoint specifications (OpenAPI/Swagger)
 ├── data-contracts.md           # Request/response schemas and validation rules
@@ -531,7 +533,7 @@ Upon completion, update `workflow-session.json`:
       "api_designer": {
         "status": "completed",
         "completed_at": "timestamp",
-        "output_directory": ".workflow/WFS-{topic}/.brainstorming/api-designer/",
+        "output_directory": ".workflow/active/WFS-{topic}/.brainstorming/api-designer/",
         "key_insights": ["endpoint_design", "versioning_strategy", "data_contracts"]
       }
     }

.claude/commands/workflow/brainstorm/artifacts.md

@@ -1,455 +1,363 @@
 ---
 name: artifacts
-description: Interactive clarification generating confirmed guidance specification
+description: Interactive clarification generating confirmed guidance specification through role-based analysis and synthesis
 argument-hint: "topic or challenge description [--count N]"
-allowed-tools: TodoWrite(*), Read(*), Write(*), Glob(*)
+allowed-tools: TodoWrite(*), Read(*), Write(*), Glob(*), AskUserQuestion(*)
 ---
 
 ## Overview
 
-Six-phase workflow: **Automatic project context collection** → Extract topic challenges → Select roles → Generate task-specific questions → Detect conflicts → Generate confirmed guidance (declarative statements only).
+Seven-phase workflow: **Context collection** → **Topic analysis** → **Role selection** → **Role questions** → **Conflict resolution** → **Final check** → **Generate specification**
+
+All user interactions use AskUserQuestion tool (max 4 questions per call, multi-round).
 
 **Input**: `"GOAL: [objective] SCOPE: [boundaries] CONTEXT: [background]" [--count N]`
-**Output**: `.workflow/WFS-{topic}/.brainstorming/guidance-specification.md` (CONFIRMED/SELECTED format)
-**Core Principle**: Questions dynamically generated from project context + topic keywords/challenges, NOT from generic templates
+**Output**: `.workflow/active/WFS-{topic}/.brainstorming/guidance-specification.md`
+**Core Principle**: Questions dynamically generated from project context + topic keywords, NOT generic templates
 
 **Parameters**:
 - `topic` (required): Topic or challenge description (structured format recommended)
-- `--count N` (optional): Number of roles user WANTS to select (system will recommend N+2 options for user to choose from, default: 3)
+- `--count N` (optional): Number of roles to select (system recommends N+2 options, default: 3)
+
+---
+
+## Quick Reference
+
+### Phase Summary
+
+| Phase | Goal | AskUserQuestion | Storage |
+|-------|------|-----------------|---------|
+| 0 | Context collection | - | context-package.json |
+| 1 | Topic analysis | 2-4 questions | intent_context |
+| 2 | Role selection | 1 multi-select | selected_roles |
+| 3 | Role questions | 3-4 per role | role_decisions[role] |
+| 4 | Conflict resolution | max 4 per round | cross_role_decisions |
+| 4.5 | Final check | progressive rounds | additional_decisions |
+| 5 | Generate spec | - | guidance-specification.md |
+
+### AskUserQuestion Pattern
+
+```javascript
+// Single-select (Phase 1, 3, 4)
+AskUserQuestion({
+  questions: [
+    {
+      question: "{问题文本}",
+      header: "{短标签}",  // max 12 chars
+      multiSelect: false,
+      options: [
+        { label: "{选项}", description: "{说明和影响}" },
+        { label: "{选项}", description: "{说明和影响}" },
+        { label: "{选项}", description: "{说明和影响}" }
+      ]
+    }
+    // ... max 4 questions per call
+  ]
+})
+
+// Multi-select (Phase 2)
+AskUserQuestion({
+  questions: [{
+    question: "请选择 {count} 个角色",
+    header: "角色选择",
+    multiSelect: true,
+    options: [/* max 4 options per call */]
+  }]
+})
+```
+
+### Multi-Round Execution
+
+```javascript
+const BATCH_SIZE = 4;
+for (let i = 0; i < allQuestions.length; i += BATCH_SIZE) {
+  const batch = allQuestions.slice(i, i + BATCH_SIZE);
+  AskUserQuestion({ questions: batch });
+  // Store responses before next round
+}
+```
+
+---
 
 ## Task Tracking
 
-**⚠️ TodoWrite Rule**: EXTEND auto-parallel's task list (NOT replace/overwrite)
+**TodoWrite Rule**: EXTEND auto-parallel's task list (NOT replace/overwrite)
 
 **When called from auto-parallel**:
-- Find the artifacts parent task: "Execute artifacts command for interactive framework generation"
-- Mark parent task as "in_progress"
-- APPEND artifacts sub-tasks AFTER the parent task (Phase 0-5)
-- Mark each sub-task as it completes
-- When Phase 5 completes, mark parent task as "completed"
-- **PRESERVE all other auto-parallel tasks** (role agents, synthesis)
+- Find artifacts parent task → Mark "in_progress"
+- APPEND sub-tasks (Phase 0-5) → Mark each as completes
+- When Phase 5 completes → Mark parent "completed"
+- PRESERVE all other auto-parallel tasks
 
 **Standalone Mode**:
 ```json
 [
-  {"content": "Initialize session (.workflow/.active-* check, parse --count parameter)", "status": "pending", "activeForm": "Initializing"},
-  {"content": "Phase 0: Automatic project context collection (call context-gather)", "status": "pending", "activeForm": "Phase 0 context collection"},
-  {"content": "Phase 1: Extract challenges, output 2-4 task-specific questions, wait for user input", "status": "pending", "activeForm": "Phase 1 topic analysis"},
-  {"content": "Phase 2: Recommend count+2 roles, output role selection, wait for user input", "status": "pending", "activeForm": "Phase 2 role selection"},
-  {"content": "Phase 3: Generate 3-4 questions per role, output and wait for answers (max 10 per round)", "status": "pending", "activeForm": "Phase 3 role questions"},
-  {"content": "Phase 4: Detect conflicts, output clarifications, wait for answers (max 10 per round)", "status": "pending", "activeForm": "Phase 4 conflict resolution"},
-  {"content": "Phase 5: Transform Q&A to declarative statements, write guidance-specification.md", "status": "pending", "activeForm": "Phase 5 document generation"}
+  {"content": "Initialize session", "status": "pending", "activeForm": "Initializing"},
+  {"content": "Phase 0: Context collection", "status": "pending", "activeForm": "Phase 0"},
+  {"content": "Phase 1: Topic analysis (2-4 questions)", "status": "pending", "activeForm": "Phase 1"},
+  {"content": "Phase 2: Role selection", "status": "pending", "activeForm": "Phase 2"},
+  {"content": "Phase 3: Role questions (per role)", "status": "pending", "activeForm": "Phase 3"},
+  {"content": "Phase 4: Conflict resolution", "status": "pending", "activeForm": "Phase 4"},
+  {"content": "Phase 4.5: Final clarification", "status": "pending", "activeForm": "Phase 4.5"},
+  {"content": "Phase 5: Generate specification", "status": "pending", "activeForm": "Phase 5"}
 ]
 ```
 
-## User Interaction Protocol
-
-### Question Output Format
-
-All questions output as structured text (detailed format with descriptions):
-
-```markdown
-【问题{N} - {短标签}】{问题文本}
-a) {选项标签}
-   说明：{选项说明和影响}
-b) {选项标签}
-   说明：{选项说明和影响}
-c) {选项标签}
-   说明：{选项说明和影响}
-
-请回答：{N}a 或 {N}b 或 {N}c
-```
-
-**Multi-select format** (Phase 2 role selection):
-```markdown
-【角色选择】请选择 {count} 个角色参与头脑风暴分析
-
-a) {role-name} ({中文名})
-   推荐理由：{基于topic的相关性说明}
-b) {role-name} ({中文名})
-   推荐理由：{基于topic的相关性说明}
-...
-
-支持格式：
-- 分别选择：2a 2c 2d (选择第2题的a、c、d选项)
-- 合并语法：2acd (选择a、c、d)
-- 逗号分隔：2a,c,d
-
-请输入选择：
-```
-
-### Input Parsing Rules
-
-**Supported formats** (intelligent parsing):
-
-1. **Space-separated**: `1a 2b 3c` → Q1:a, Q2:b, Q3:c
-2. **Comma-separated**: `1a,2b,3c` → Q1:a, Q2:b, Q3:c
-3. **Multi-select combined**: `2abc` → Q2: options a,b,c
-4. **Multi-select spaces**: `2 a b c` → Q2: options a,b,c
-5. **Multi-select comma**: `2a,b,c` → Q2: options a,b,c
-6. **Natural language**: `问题1选a` → 1a (fallback parsing)
-
-**Parsing algorithm**:
-- Extract question numbers and option letters
-- Validate question numbers match output
-- Validate option letters exist for each question
-- If ambiguous/invalid, output example format and request re-input
-
-**Error handling** (lenient):
-- Recognize common variations automatically
-- If parsing fails, show example and wait for clarification
-- Support re-input without penalty
-
-### Batching Strategy
-
-**Batch limits**:
-- **Default**: Maximum 10 questions per round
-- **Phase 2 (role selection)**: Display all recommended roles at once (count+2 roles)
-- **Auto-split**: If questions > 10, split into multiple rounds with clear round indicators
-
-**Round indicators**:
-```markdown
-===== 第 1 轮问题 (共2轮) =====
-【问题1 - ...】...
-【问题2 - ...】...
-...
-【问题10 - ...】...
-
-请回答 (格式: 1a 2b ... 10c)：
-```
-
-### Interaction Flow
-
-**Standard flow**:
-1. Output questions in formatted text
-2. Output expected input format example
-3. Wait for user input
-4. Parse input with intelligent matching
-5. If parsing succeeds → Store answers and continue
-6. If parsing fails → Show error, example, and wait for re-input
-
-**No question/option limits**: Text-based interaction removes previous 4-question and 4-option restrictions
+---
 
 ## Execution Phases
 
 ### Session Management
-- Check `.workflow/.active-*` markers first
-- Multiple sessions → Prompt selection | Single → Use it | None → Create `WFS-[topic-slug]`
-- Parse `--count N` parameter from user input (default: 3 if not specified)
-- Store decisions in `workflow-session.json` including count parameter
 
-### Phase 0: Automatic Project Context Collection
+- Check `.workflow/active/` for existing sessions
+- Multiple → Prompt selection | Single → Use it | None → Create `WFS-[topic-slug]`
+- Parse `--count N` parameter (default: 3)
+- Store decisions in `workflow-session.json`
 
-**Goal**: Gather project architecture, documentation, and relevant code context BEFORE user interaction
+### Phase 0: Context Collection
 
-**Detection Mechanism** (execute first):
-```javascript
-// Check if context-package already exists
-const contextPackagePath = `.workflow/WFS-{session-id}/.process/context-package.json`;
+**Goal**: Gather project context BEFORE user interaction
 
-if (file_exists(contextPackagePath)) {
-  // Validate package
-  const package = Read(contextPackagePath);
-  if (package.metadata.session_id === session_id) {
-    console.log("✅ Valid context-package found, skipping Phase 0");
-    return; // Skip to Phase 1
-  }
-}
-```
+**Steps**:
+1. Check if `context-package.json` exists → Skip if valid
+2. Invoke `context-search-agent` (BRAINSTORM MODE - lightweight)
+3. Output: `.workflow/active/WFS-{session-id}/.process/context-package.json`
 
-**Implementation**: Invoke `context-search-agent` only if package doesn't exist
+**Graceful Degradation**: If agent fails, continue to Phase 1 without context
 
 ```javascript
 Task(
   subagent_type="context-search-agent",
+  run_in_background=false,
   description="Gather project context for brainstorm",
   prompt=`
-You are executing as context-search-agent (.claude/agents/context-search-agent.md).
+Execute context-search-agent in BRAINSTORM MODE (Phase 1-2 only).
 
-## Execution Mode
-**BRAINSTORM MODE** (Lightweight) - Phase 1-2 only (skip deep analysis)
+Session: ${session_id}
+Task: ${task_description}
+Output: .workflow/${session_id}/.process/context-package.json
 
-## Session Information
-- **Session ID**: ${session_id}
-- **Task Description**: ${task_description}
-- **Output Path**: .workflow/${session_id}/.process/context-package.json
-
-## Mission
-Execute complete context-search-agent workflow for implementation planning:
-
-### Phase 1: Initialization & Pre-Analysis
-1. **Detection**: Check for existing context-package (early exit if valid)
-2. **Foundation**: Initialize code-index, get project structure, load docs
-3. **Analysis**: Extract keywords, determine scope, classify complexity
-
-### Phase 2: Multi-Source Context Discovery
-Execute all 3 discovery tracks:
-- **Track 1**: Reference documentation (CLAUDE.md, architecture docs)
-- **Track 2**: Web examples (use Exa MCP for unfamiliar tech/APIs)
-- **Track 3**: Codebase analysis (5-layer discovery: files, content, patterns, deps, config/tests)
-
-### Phase 3: Synthesis, Assessment & Packaging
-1. Apply relevance scoring and build dependency graph
-2. Synthesize 3-source data (docs > code > web)
-3. Integrate brainstorm artifacts (if .brainstorming/ exists, read content)
-4. Perform conflict detection with risk assessment
-5. Generate and validate context-package.json
-
-## Output Requirements
-Complete context-package.json with:
-- **metadata**: task_description, keywords, complexity, tech_stack, session_id
-- **project_context**: architecture_patterns, coding_conventions, tech_stack
-- **assets**: {documentation[], source_code[], config[], tests[]} with relevance scores
-- **dependencies**: {internal[], external[]} with dependency graph
-- **brainstorm_artifacts**: {guidance_specification, role_analyses[], synthesis_output} with content
-- **conflict_detection**: {risk_level, risk_factors, affected_modules[], mitigation_strategy}
-
-## Quality Validation
-Before completion verify:
-- [ ] Valid JSON format with all required fields
-- [ ] File relevance accuracy >80%
-- [ ] Dependency graph complete (max 2 transitive levels)
-- [ ] Conflict risk level calculated correctly
-- [ ] No sensitive data exposed
-- [ ] Total files ≤50 (prioritize high-relevance)
-
-Execute autonomously following agent documentation.
-Report completion with statistics.
+Required fields: metadata, project_context, assets, dependencies, conflict_detection
 `
 )
 ```
 
-**Graceful Degradation**:
-- If agent fails: Log warning, continue to Phase 1 without project context
-- If package invalid: Re-run context-search-agent
+### Phase 1: Topic Analysis
 
-### Phase 1: Topic Analysis & Intent Classification
-
-**Goal**: Extract keywords/challenges to drive all subsequent question generation, **enriched by Phase 0 project context**
+**Goal**: Extract keywords/challenges enriched by Phase 0 context
 
 **Steps**:
-1. **Load Phase 0 context** (if available):
-   - Read `.workflow/WFS-{session-id}/.process/context-package.json`
-   - Extract: tech_stack, existing modules, conflict_risk, relevant files
+1. Load Phase 0 context (tech_stack, modules, conflict_risk)
+2. Deep topic analysis (entities, challenges, constraints, metrics)
+3. Generate 2-4 context-aware probing questions
+4. AskUserQuestion → Store to `session.intent_context`
 
-2. **Deep topic analysis** (context-aware):
-   - Extract technical entities from topic + existing codebase
-   - Identify core challenges considering existing architecture
-   - Consider constraints (timeline/budget/compliance)
-   - Define success metrics based on current project state
-
-3. **Generate 2-4 context-aware probing questions**:
-   - Reference existing tech stack in questions
-   - Consider integration with existing modules
-   - Address identified conflict risks from Phase 0
-   - Target root challenges and trade-off priorities
-
-4. **User interaction**: Output questions using text format (see User Interaction Protocol), wait for user input
-
-5. **Parse user answers**: Use intelligent parsing to extract answers from user input (support multiple formats)
-
-6. **Storage**: Store answers to `session.intent_context` with `{extracted_keywords, identified_challenges, user_answers, project_context_used}`
-
-**Example Output**:
-```markdown
-===== Phase 1: 项目意图分析 =====
-
-【问题1 - 核心挑战】实时协作平台的主要技术挑战？
-a) 实时数据同步
-   说明：100+用户同时在线，状态同步复杂度高
-b) 可扩展性架构
-   说明：用户规模增长时的系统扩展能力
-c) 冲突解决机制
-   说明：多用户同时编辑的冲突处理策略
-
-【问题2 - 优先级】MVP阶段最关注的指标？
-a) 功能完整性
-   说明：实现所有核心功能
-b) 用户体验
-   说明：流畅的交互体验和响应速度
-c) 系统稳定性
-   说明：高可用性和数据一致性
-
-请回答 (格式: 1a 2b)：
+**Example**:
+```javascript
+AskUserQuestion({
+  questions: [
+    {
+      question: "实时协作平台的主要技术挑战？",
+      header: "核心挑战",
+      multiSelect: false,
+      options: [
+        { label: "实时数据同步", description: "100+用户同时在线，状态同步复杂度高" },
+        { label: "可扩展性架构", description: "用户规模增长时的系统扩展能力" },
+        { label: "冲突解决机制", description: "多用户同时编辑的冲突处理策略" }
+      ]
+    },
+    {
+      question: "MVP阶段最关注的指标？",
+      header: "优先级",
+      multiSelect: false,
+      options: [
+        { label: "功能完整性", description: "实现所有核心功能" },
+        { label: "用户体验", description: "流畅的交互体验和响应速度" },
+        { label: "系统稳定性", description: "高可用性和数据一致性" }
+      ]
+    }
+  ]
+})
 ```
 
-**User input examples**:
-- `1a 2c` → Q1:a, Q2:c
-- `1a,2c` → Q1:a, Q2:c
-
 **⚠️ CRITICAL**: Questions MUST reference topic keywords. Generic "Project type?" violates dynamic generation.
 
 ### Phase 2: Role Selection
 
-**⚠️ CRITICAL**: User MUST interact to select roles. NEVER auto-select without user confirmation.
+**Goal**: User selects roles from intelligent recommendations
 
-**Available Roles**:
-- data-architect (数据架构师)
-- product-manager (产品经理)
-- product-owner (产品负责人)
-- scrum-master (敏捷教练)
-- subject-matter-expert (领域专家)
-- system-architect (系统架构师)
-- test-strategist (测试策略师)
-- ui-designer (UI 设计师)
-- ux-expert (UX 专家)
+**Available Roles**: data-architect, product-manager, product-owner, scrum-master, subject-matter-expert, system-architect, test-strategist, ui-designer, ux-expert
 
 **Steps**:
-1. **Intelligent role recommendation** (AI analysis):
-   - Analyze Phase 1 extracted keywords and challenges
-   - Use AI reasoning to determine most relevant roles for the specific topic
-   - Recommend count+2 roles (e.g., if user wants 3 roles, recommend 5 options)
-   - Provide clear rationale for each recommended role based on topic context
+1. Analyze Phase 1 keywords → Recommend count+2 roles with rationale
+2. AskUserQuestion (multiSelect=true) → Store to `session.selected_roles`
+3. If count+2 > 4, split into multiple rounds
 
-2. **User selection** (text interaction):
-   - Output all recommended roles at once (no batching needed for count+2 roles)
-   - Display roles with labels and relevance rationale
-   - Wait for user input in multi-select format
-   - Parse user input (support multiple formats)
-   - **Storage**: Store selections to `session.selected_roles`
-
-**Example Output**:
-```markdown
-===== Phase 2: 角色选择 =====
-
-【角色选择】请选择 3 个角色参与头脑风暴分析
-
-a) system-architect (系统架构师)
-   推荐理由：实时同步架构设计和技术选型的核心角色
-b) ui-designer (UI设计师)
-   推荐理由：协作界面用户体验和实时状态展示
-c) product-manager (产品经理)
-   推荐理由：功能优先级和MVP范围决策
-d) data-architect (数据架构师)
-   推荐理由：数据同步模型和存储方案设计
-e) ux-expert (UX专家)
-   推荐理由：多用户协作交互流程优化
-
-支持格式：
-- 分别选择：2a 2c 2d (选择a、c、d)
-- 合并语法：2acd (选择a、c、d)
-- 逗号分隔：2a,c,d (选择a、c、d)
-
-请输入选择：
+**Example**:
+```javascript
+AskUserQuestion({
+  questions: [{
+    question: "请选择 3 个角色参与头脑风暴分析",
+    header: "角色选择",
+    multiSelect: true,
+    options: [
+      { label: "system-architect", description: "实时同步架构设计和技术选型" },
+      { label: "ui-designer", description: "协作界面用户体验和状态展示" },
+      { label: "product-manager", description: "功能优先级和MVP范围决策" },
+      { label: "data-architect", description: "数据同步模型和存储方案设计" }
+    ]
+  }]
+})
 ```
 
-**User input examples**:
-- `2acd` → Roles: a, c, d (system-architect, product-manager, data-architect)
-- `2a 2c 2d` → Same result
-- `2a,c,d` → Same result
+**⚠️ CRITICAL**: User MUST interact. NEVER auto-select without confirmation.
 
-**Role Recommendation Rules**:
-- NO hardcoded keyword-to-role mappings
-- Use intelligent analysis of topic, challenges, and requirements
-- Consider role synergies and coverage gaps
-- Explain WHY each role is relevant to THIS specific topic
-- Default recommendation: count+2 roles for user to choose from
-
-### Phase 3: Role-Specific Questions (Dynamic Generation)
+### Phase 3: Role-Specific Questions
 
 **Goal**: Generate deep questions mapping role expertise to Phase 1 challenges
 
 **Algorithm**:
-```
-FOR each selected role:
-  1. Map Phase 1 challenges to role domain:
-     - "real-time sync" + system-architect → State management pattern
-     - "100 users" + system-architect → Communication protocol
-     - "low latency" + system-architect → Conflict resolution
+1. FOR each selected role:
+   - Map Phase 1 challenges to role domain
+   - Generate 3-4 questions (implementation depth, trade-offs, edge cases)
+   - AskUserQuestion per role → Store to `session.role_decisions[role]`
+2. Process roles sequentially (one at a time for clarity)
+3. If role needs > 4 questions, split into multiple rounds
 
-  2. Generate 3-4 questions per role probing implementation depth, trade-offs, edge cases:
-     Q: "How handle real-time state sync for 100+ users?" (explores approach)
-     Q: "How resolve conflicts when 2 users edit simultaneously?" (explores edge case)
-     Options: [Event Sourcing/Centralized/CRDT] (concrete, explain trade-offs for THIS use case)
-
-  3. Output questions in text format per role:
-     - Display all questions for current role (3-4 questions, no 10-question limit)
-     - Questions in Chinese (用中文提问)
-     - Wait for user input
-     - Parse answers using intelligent parsing
-     - Store answers to session.role_decisions[role]
+**Example** (system-architect):
+```javascript
+AskUserQuestion({
+  questions: [
+    {
+      question: "100+ 用户实时状态同步方案？",
+      header: "状态同步",
+      multiSelect: false,
+      options: [
+        { label: "Event Sourcing", description: "完整事件历史，支持回溯，存储成本高" },
+        { label: "集中式状态管理", description: "实现简单，单点瓶颈风险" },
+        { label: "CRDT", description: "去中心化，自动合并，学习曲线陡" }
+      ]
+    },
+    {
+      question: "两个用户同时编辑冲突如何解决？",
+      header: "冲突解决",
+      multiSelect: false,
+      options: [
+        { label: "自动合并", description: "用户无感知，可能产生意外结果" },
+        { label: "手动解决", description: "用户控制，增加交互复杂度" },
+        { label: "版本控制", description: "保留历史，需要分支管理" }
+      ]
+    }
+  ]
+})
 ```
 
-**Batching Strategy**:
-- Each role outputs all its questions at once (typically 3-4 questions)
-- No need to split per role (within 10-question batch limit)
-- Multiple roles processed sequentially (one role at a time for clarity)
+### Phase 4: Conflict Resolution
 
-**Output Format**: Follow standard format from "User Interaction Protocol" section (single-choice question format)
-
-**Example Topic-Specific Questions** (system-architect role for "real-time collaboration platform"):
-- "100+ 用户实时状态同步方案?" → Options: Event Sourcing / 集中式状态管理 / CRDT
-- "两个用户同时编辑冲突如何解决?" → Options: 自动合并 / 手动解决 / 版本控制
-- "低延迟通信协议选择?" → Options: WebSocket / SSE / 轮询
-- "系统扩展性架构方案?" → Options: 微服务 / 单体+缓存 / Serverless
-
-**Quality Requirements**: See "Question Generation Guidelines" section for detailed rules
-
-### Phase 4: Cross-Role Clarification (Conflict Detection)
-
-**Goal**: Resolve ACTUAL conflicts from Phase 3 answers, not pre-defined relationships
+**Goal**: Resolve ACTUAL conflicts from Phase 3 answers
 
 **Algorithm**:
-```
 1. Analyze Phase 3 answers for conflicts:
-   - Contradictory choices: product-manager "fast iteration" vs system-architect "complex Event Sourcing"
-   - Missing integration: ui-designer "Optimistic updates" but system-architect didn't address conflict handling
-   - Implicit dependencies: ui-designer "Live cursors" but no auth approach defined
-
-2. FOR each detected conflict:
-   Generate clarification questions referencing SPECIFIC Phase 3 choices
-
-3. Output clarification questions in text format:
-   - Batch conflicts into rounds (max 10 questions per round)
-   - Display questions with context from Phase 3 answers
-   - Questions in Chinese (用中文提问)
-   - Wait for user input
-   - Parse answers using intelligent parsing
-   - Store answers to session.cross_role_decisions
-
+   - Contradictory choices (e.g., "fast iteration" vs "complex Event Sourcing")
+   - Missing integration (e.g., "Optimistic updates" but no conflict handling)
+   - Implicit dependencies (e.g., "Live cursors" but no auth defined)
+2. Generate clarification questions referencing SPECIFIC Phase 3 choices
+3. AskUserQuestion (max 4 per call, multi-round) → Store to `session.cross_role_decisions`
 4. If NO conflicts: Skip Phase 4 (inform user: "未检测到跨角色冲突，跳过Phase 4")
+
+**Example**:
+```javascript
+AskUserQuestion({
+  questions: [{
+    question: "CRDT 与 UI 回滚期望冲突，如何解决？\n背景：system-architect选择CRDT，ui-designer期望回滚UI",
+    header: "架构冲突",
+    multiSelect: false,
+    options: [
+      { label: "采用 CRDT", description: "保持去中心化，调整UI期望" },
+      { label: "显示合并界面", description: "增加用户交互，展示冲突详情" },
+      { label: "切换到 OT", description: "支持回滚，增加服务器复杂度" }
+    ]
+  }]
+})
 ```
 
-**Batching Strategy**:
-- Maximum 10 clarification questions per round
-- If conflicts > 10, split into multiple rounds
-- Prioritize most critical conflicts first
+### Phase 4.5: Final Clarification
 
-**Output Format**: Follow standard format from "User Interaction Protocol" section (single-choice question format with background context)
-
-**Example Conflict Detection** (from Phase 3 answers):
-- **Architecture Conflict**: "CRDT 与 UI 回滚期望冲突，如何解决?"
-  - Background: system-architect chose CRDT, ui-designer expects rollback UI
-  - Options: 采用 CRDT / 显示合并界面 / 切换到 OT
-- **Integration Gap**: "实时光标功能缺少身份认证方案"
-  - Background: ui-designer chose live cursors, no auth defined
-  - Options: OAuth 2.0 / JWT Token / Session-based
-
-**Quality Requirements**: See "Question Generation Guidelines" section for conflict-specific rules
-
-### Phase 5: Generate Guidance Specification
+**Purpose**: Ensure no important points missed before generating specification
 
 **Steps**:
-1. Load all decisions: `intent_context` + `selected_roles` + `role_decisions` + `cross_role_decisions`
-2. Transform Q&A pairs to declarative: Questions → Headers, Answers → CONFIRMED/SELECTED statements
-3. Generate guidance-specification.md (template below) - **PRIMARY OUTPUT FILE**
-4. Update workflow-session.json with **METADATA ONLY**:
-   - session_id (e.g., "WFS-topic-slug")
-   - selected_roles[] (array of role names, e.g., ["system-architect", "ui-designer", "product-manager"])
-   - topic (original user input string)
-   - timestamp (ISO-8601 format)
-   - phase_completed: "artifacts"
-   - count_parameter (number from --count flag)
-5. Validate: No interrogative sentences in .md file, all decisions traceable, no content duplication in .json
+1. Ask initial check:
+   ```javascript
+   AskUserQuestion({
+     questions: [{
+       question: "在生成最终规范之前，是否有前面未澄清的重点需要补充？",
+       header: "补充确认",
+       multiSelect: false,
+       options: [
+         { label: "无需补充", description: "前面的讨论已经足够完整" },
+         { label: "需要补充", description: "还有重要内容需要澄清" }
+       ]
+     }]
+   })
+   ```
+2. If "需要补充":
+   - Analyze user's additional points
+   - Generate progressive questions (not role-bound, interconnected)
+   - AskUserQuestion (max 4 per round) → Store to `session.additional_decisions`
+   - Repeat until user confirms completion
+3. If "无需补充": Proceed to Phase 5
 
-**⚠️ CRITICAL OUTPUT SEPARATION**:
-- **guidance-specification.md**: Full guidance content (decisions, rationale, integration points)
-- **workflow-session.json**: Session metadata ONLY (no guidance content, no decisions, no Q&A pairs)
-- **NO content duplication**: Guidance stays in .md, metadata stays in .json
+**Progressive Pattern**: Questions interconnected, each round informs next, continue until resolved.
 
-## Output Document Template
+### Phase 5: Generate Specification
 
-**File**: `.workflow/WFS-{topic}/.brainstorming/guidance-specification.md`
+**Steps**:
+1. Load all decisions: `intent_context` + `selected_roles` + `role_decisions` + `cross_role_decisions` + `additional_decisions`
+2. Transform Q&A to declarative: Questions → Headers, Answers → CONFIRMED/SELECTED statements
+3. Generate `guidance-specification.md`
+4. Update `workflow-session.json` (metadata only)
+5. Validate: No interrogative sentences, all decisions traceable
+
+---
+
+## Question Guidelines
+
+### Core Principle
+
+**Target**: 开发者（理解技术但需要从用户需求出发）
+
+**Question Structure**: `[业务场景/需求前提] + [技术关注点]`
+**Option Structure**: `标签：[技术方案] + 说明：[业务影响] + [技术权衡]`
+
+### Quality Rules
+
+**MUST Include**:
+- ✅ All questions in Chinese (用中文提问)
+- ✅ 业务场景作为问题前提
+- ✅ 技术选项的业务影响说明
+- ✅ 量化指标和约束条件
+
+**MUST Avoid**:
+- ❌ 纯技术选型无业务上下文
+- ❌ 过度抽象的用户体验问题
+- ❌ 脱离话题的通用架构问题
+
+### Phase-Specific Requirements
+
+| Phase | Focus | Key Requirements |
+|-------|-------|------------------|
+| 1 | 意图理解 | Reference topic keywords, 用户场景、业务约束、优先级 |
+| 2 | 角色推荐 | Intelligent analysis (NOT keyword mapping), explain relevance |
+| 3 | 角色问题 | Reference Phase 1 keywords, concrete options with trade-offs |
+| 4 | 冲突解决 | Reference SPECIFIC Phase 3 choices, explain impact on both roles |
+
+---
+
+## Output & Governance
+
+### Output Template
+
+**File**: `.workflow/active/WFS-{topic}/.brainstorming/guidance-specification.md`
 
 ```markdown
 # [Project] - Confirmed Guidance Specification
@@ -478,9 +386,9 @@ FOR each selected role:
 
 ## Next Steps
 **⚠️ Automatic Continuation** (when called from auto-parallel):
-- auto-parallel will assign agents to generate role-specific analysis documents
-- Each selected role gets dedicated conceptual-planning-agent
-- Agents read this guidance-specification.md for framework context
+- auto-parallel assigns agents for role-specific analysis
+- Each selected role gets conceptual-planning-agent
+- Agents read this guidance-specification.md for context
 
 ## Appendix: Decision Tracking
 | Decision ID | Category | Question | Selected | Phase | Rationale |
@@ -490,95 +398,19 @@ FOR each selected role:
 | D-003+ | [Role] | [Q] | [A] | 3 | [Why] |
 ```
 
-## Question Generation Guidelines
-
-### Core Principle: Developer-Facing Questions with User Context
-
-**Target Audience**: 开发者（理解技术但需要从用户需求出发）
-
-**Generation Philosophy**:
-1. **Phase 1**: 用户场景、业务约束、优先级（建立上下文）
-2. **Phase 2**: 基于话题分析的智能角色推荐（非关键词映射）
-3. **Phase 3**: 业务需求 + 技术选型（需求驱动的技术决策）
-4. **Phase 4**: 技术冲突的业务权衡（帮助开发者理解影响）
-
-### Universal Quality Rules
-
-**Question Structure** (all phases):
-```
-[业务场景/需求前提] + [技术关注点]
-```
-
-**Option Structure** (all phases):
-```
-标签：[技术方案简称] + (业务特征)
-说明：[业务影响] + [技术权衡]
-```
-
-**MUST Include** (all phases):
-- ✅ All questions in Chinese (用中文提问)
-- ✅ 业务场景作为问题前提
-- ✅ 技术选项的业务影响说明
-- ✅ 量化指标和约束条件
-
-**MUST Avoid** (all phases):
-- ❌ 纯技术选型无业务上下文
-- ❌ 过度抽象的用户体验问题
-- ❌ 脱离话题的通用架构问题
-
-### Phase-Specific Requirements
-
-**Phase 1 Requirements**:
-- Questions MUST reference topic keywords (NOT generic "Project type?")
-- Focus: 用户使用场景（谁用？怎么用？多频繁？）、业务约束（预算、时间、团队、合规）
-- Success metrics: 性能指标、用户体验目标
-- Priority ranking: MVP vs 长期规划
-
-**Phase 3 Requirements**:
-- Questions MUST reference Phase 1 keywords (e.g., "real-time", "100 users")
-- Options MUST be concrete approaches with relevance to topic
-- Each option includes trade-offs specific to this use case
-- Include 业务需求驱动的技术问题、量化指标（并发数、延迟、可用性）
-
-**Phase 4 Requirements**:
-- Questions MUST reference SPECIFIC Phase 3 choices in background context
-- Options address the detected conflict directly
-- Each option explains impact on both conflicting roles
-- NEVER use static "Cross-Role Matrix" - ALWAYS analyze actual Phase 3 answers
-- Focus: 技术冲突的业务权衡、帮助开发者理解不同选择的影响
-
-## Validation Checklist
-
-Generated guidance-specification.md MUST:
-- ✅ No interrogative sentences (use CONFIRMED/SELECTED)
-- ✅ Every decision traceable to user answer
-- ✅ Cross-role conflicts resolved or documented
-- ✅ Next steps concrete and specific
-- ✅ All Phase 1-4 decisions in session metadata
-
-## Update Mechanism
+### File Structure
 
 ```
-IF guidance-specification.md EXISTS:
-  Prompt: "Regenerate completely / Update sections / Cancel"
-ELSE:
-  Run full Phase 1-5 flow
+.workflow/active/WFS-[topic]/
+├── workflow-session.json              # Metadata ONLY
+├── .process/
+│   └── context-package.json           # Phase 0 output
+└── .brainstorming/
+    └── guidance-specification.md      # Full guidance content
 ```
 
-## Governance Rules
+### Session Metadata
 
-**Output Requirements**:
-- All decisions MUST use CONFIRMED/SELECTED (NO "?" in decision sections)
-- Every decision MUST trace to user answer
-- Conflicts MUST be resolved (not marked "TBD")
-- Next steps MUST be actionable
-- Topic preserved as authoritative reference in session
-
-**CRITICAL**: Guidance is single source of truth for downstream phases. Ambiguity violates governance.
-
-## Storage Validation
-
-**workflow-session.json** (metadata only):
 ```json
 {
   "session_id": "WFS-{topic-slug}",
@@ -591,15 +423,31 @@ ELSE:
 }
 ```
 
-**⚠️ Rule**: Session JSON stores ONLY metadata (session_id, selected_roles[], topic, timestamps). All guidance content goes to guidance-specification.md.
+**⚠️ Rule**: Session JSON stores ONLY metadata. All guidance content goes to guidance-specification.md.
 
-## File Structure
+### Validation Checklist
+
+- ✅ No interrogative sentences (use CONFIRMED/SELECTED)
+- ✅ Every decision traceable to user answer
+- ✅ Cross-role conflicts resolved or documented
+- ✅ Next steps concrete and specific
+- ✅ No content duplication between .json and .md
+
+### Update Mechanism
 
 ```
-.workflow/WFS-[topic]/
-├── .active-brainstorming
-├── workflow-session.json              # Session metadata ONLY
-└── .brainstorming/
-    └── guidance-specification.md      # Full guidance content
+IF guidance-specification.md EXISTS:
+  Prompt: "Regenerate completely / Update sections / Cancel"
+ELSE:
+  Run full Phase 0-5 flow
 ```
 
+### Governance Rules
+
+- All decisions MUST use CONFIRMED/SELECTED (NO "?" in decision sections)
+- Every decision MUST trace to user answer
+- Conflicts MUST be resolved (not marked "TBD")
+- Next steps MUST be actionable
+- Topic preserved as authoritative reference
+
+**CRITICAL**: Guidance is single source of truth for downstream phases. Ambiguity violates governance.

.claude/commands/workflow/brainstorm/auto-parallel.md

@@ -1,6 +1,6 @@
 ---
 name: auto-parallel
-description: Parallel brainstorming automation with dynamic role selection and concurrent execution
+description: Parallel brainstorming automation with dynamic role selection and concurrent execution across multiple perspectives
 argument-hint: "topic or challenge description" [--count N]
 allowed-tools: SlashCommand(*), Task(*), TodoWrite(*), Read(*), Write(*), Bash(*), Glob(*)
 ---
@@ -9,52 +9,69 @@ allowed-tools: SlashCommand(*), Task(*), TodoWrite(*), Read(*), Write(*), Bash(*
 
 ## Coordinator Role
 
-**This command is a pure orchestrator**: Execute 3 phases in sequence (interactive framework → parallel role analysis → synthesis), delegate to specialized commands/agents, and ensure complete execution through **automatic continuation**.
+**This command is a pure orchestrator**: Dispatches 3 phases in sequence (interactive framework → parallel role analysis → synthesis), coordinating specialized commands/agents through task attachment model.
+
+**Task Attachment Model**:
+- SlashCommand dispatch **expands workflow** by attaching sub-tasks to current TodoWrite
+- Task agent dispatch **attaches analysis tasks** to orchestrator's TodoWrite
+- Phase 1: artifacts command attaches its internal tasks (Phase 1-5)
+- Phase 2: N conceptual-planning-agent tasks attached in parallel
+- Phase 3: synthesis command attaches its internal tasks
+- Orchestrator **executes these attached tasks** sequentially (Phase 1, 3) or in parallel (Phase 2)
+- After completion, attached tasks are **collapsed** back to high-level phase summary
+- This is **task expansion**, not external delegation
 
 **Execution Model - Auto-Continue Workflow**:
 
 This workflow runs **fully autonomously** once triggered. Phase 1 (artifacts) handles user interaction, Phase 2 (role agents) runs in parallel.
 
 1. **User triggers**: `/workflow:brainstorm:auto-parallel "topic" [--count N]`
-2. **Phase 1 executes** → artifacts command (interactive framework) → Auto-continues
-3. **Phase 2 executes** → Parallel role agents (N agents run concurrently) → Auto-continues
-4. **Phase 3 executes** → Synthesis command → Reports final summary
+2. **Dispatch Phase 1** → artifacts command (tasks ATTACHED) → Auto-continues
+3. **Dispatch Phase 2** → Parallel role agents (N tasks ATTACHED concurrently) → Auto-continues
+4. **Dispatch Phase 3** → Synthesis command (tasks ATTACHED) → Reports final summary
 
 **Auto-Continue Mechanism**:
-- TodoList tracks current phase status
-- After Phase 1 (artifacts) completion, automatically load roles and launch Phase 2 agents
-- After Phase 2 (all agents) completion, automatically execute Phase 3 synthesis
-- Progress updates shown at each phase for visibility
+- TodoList tracks current phase status and dynamically manages task attachment/collapse
+- When Phase 1 (artifacts) finishes executing, automatically load roles and launch Phase 2 agents
+- When Phase 2 (all agents) finishes executing, automatically execute Phase 3 synthesis
+- **⚠️ CONTINUOUS EXECUTION** - Do not stop until all phases complete
 
 ## Core Rules
 
-1. **Start Immediately**: First action is TodoWrite initialization, second action is Phase 1 command execution
+1. **Start Immediately**: First action is TodoWrite initialization, second action is dispatch Phase 1 command
 2. **No Preliminary Analysis**: Do not analyze topic before Phase 1 - artifacts handles all analysis
 3. **Parse Every Output**: Extract selected_roles from workflow-session.json after Phase 1
-4. **Auto-Continue via TodoList**: Check TodoList status to execute next pending phase automatically
-5. **Track Progress**: Update TodoWrite after every phase completion
-6. **TodoWrite Extension**: artifacts command EXTENDS parent TodoList (NOT replaces)
+4. **Auto-Continue via TodoList**: Check TodoList status to dispatch next pending phase automatically
+5. **Track Progress**: Update TodoWrite dynamically with task attachment/collapse pattern
+6. **Task Attachment Model**: SlashCommand and Task dispatches **attach** sub-tasks to current workflow. Orchestrator **executes** these attached tasks itself, then **collapses** them after completion
+7. **⚠️ CRITICAL: DO NOT STOP**: Continuous multi-phase workflow. After executing all attached tasks, immediately collapse them and dispatch next phase
+8. **Parallel Execution**: Phase 2 attaches multiple agent tasks simultaneously for concurrent execution
 
 ## Usage
 
 ```bash
-/workflow:brainstorm:auto-parallel "<topic>" [--count N]
+/workflow:brainstorm:auto-parallel "<topic>" [--count N] [--style-skill package-name]
 ```
 
 **Recommended Structured Format**:
 ```bash
-/workflow:brainstorm:auto-parallel "GOAL: [objective] SCOPE: [boundaries] CONTEXT: [background]" [--count N]
+/workflow:brainstorm:auto-parallel "GOAL: [objective] SCOPE: [boundaries] CONTEXT: [background]" [--count N] [--style-skill package-name]
 ```
 
 **Parameters**:
 - `topic` (required): Topic or challenge description (structured format recommended)
 - `--count N` (optional): Number of roles to select (default: 3, max: 9)
+- `--style-skill package-name` (optional): Style SKILL package to load for UI design (located at `.claude/skills/style-{package-name}/`)
 
 ## 3-Phase Execution
 
 ### Phase 1: Interactive Framework Generation
 
-**Command**: `SlashCommand(command="/workflow:brainstorm:artifacts \"{topic}\" --count {N}")`
+**Step 1: Dispatch** - Interactive framework generation via artifacts command
+
+```javascript
+SlashCommand(command="/workflow:brainstorm:artifacts \"{topic}\" --count {N}")
+```
 
 **What It Does**:
 - Topic analysis: Extract challenges, generate task-specific questions
@@ -72,17 +89,40 @@ This workflow runs **fully autonomously** once triggered. Phase 1 (artifacts) ha
 **Validation**:
 - guidance-specification.md created with confirmed decisions
 - workflow-session.json contains selected_roles[] (metadata only, no content duplication)
-- Session directory `.workflow/WFS-{topic}/.brainstorming/` exists
+- Session directory `.workflow/active/WFS-{topic}/.brainstorming/` exists
 
-**TodoWrite**: Mark phase 1 completed, phase 2 in_progress
+**TodoWrite Update (Phase 1 SlashCommand dispatched - tasks attached)**:
+```json
+[
+  {"content": "Phase 0: Parameter Parsing", "status": "completed", "activeForm": "Parsing count parameter"},
+  {"content": "Phase 1: Interactive Framework Generation", "status": "in_progress", "activeForm": "Executing artifacts interactive framework"},
+  {"content": "  → Topic analysis and question generation", "status": "in_progress", "activeForm": "Analyzing topic"},
+  {"content": "  → Role selection and user confirmation", "status": "pending", "activeForm": "Selecting roles"},
+  {"content": "  → Role questions and user decisions", "status": "pending", "activeForm": "Collecting role questions"},
+  {"content": "  → Conflict detection and resolution", "status": "pending", "activeForm": "Resolving conflicts"},
+  {"content": "  → Guidance specification generation", "status": "pending", "activeForm": "Generating guidance"},
+  {"content": "Phase 2: Parallel Role Analysis", "status": "pending", "activeForm": "Executing parallel role analysis"},
+  {"content": "Phase 3: Synthesis Integration", "status": "pending", "activeForm": "Executing synthesis integration"}
+]
+```
 
-**After Phase 1**: Auto-continue to Phase 2 (role agent assignment)
+**Note**: SlashCommand dispatch **attaches** artifacts' 5 internal tasks. Orchestrator **executes** these tasks sequentially.
 
-**⚠️ TodoWrite Coordination**: artifacts EXTENDS parent TodoList by:
-- Marking parent task "Execute artifacts..." as in_progress
-- APPENDING artifacts sub-tasks (Phase 1-5) after parent task
-- PRESERVING all other auto-parallel tasks (role agents, synthesis)
-- When artifacts Phase 5 completes, marking parent task as completed
+**Next Action**: Tasks attached → **Execute Phase 1.1-1.5** sequentially
+
+**TodoWrite Update (Phase 1 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Phase 0: Parameter Parsing", "status": "completed", "activeForm": "Parsing count parameter"},
+  {"content": "Phase 1: Interactive Framework Generation", "status": "completed", "activeForm": "Executing artifacts interactive framework"},
+  {"content": "Phase 2: Parallel Role Analysis", "status": "pending", "activeForm": "Executing parallel role analysis"},
+  {"content": "Phase 3: Synthesis Integration", "status": "pending", "activeForm": "Executing synthesis integration"}
+]
+```
+
+**Note**: Phase 1 tasks completed and collapsed to summary.
+
+**After Phase 1**: Auto-continue to Phase 2 (parallel role agent execution)
 
 ---
 
@@ -97,24 +137,14 @@ Execute {role-name} analysis for existing topic framework
 
 ## Context Loading
 ASSIGNED_ROLE: {role-name}
-OUTPUT_LOCATION: .workflow/WFS-{session}/.brainstorming/{role}/
+OUTPUT_LOCATION: .workflow/active/WFS-{session}/.brainstorming/{role}/
 TOPIC: {user-provided-topic}
 
 ## Flow Control Steps
-1. **load_topic_framework**
-   - Action: Load structured topic discussion framework
-   - Command: Read(.workflow/WFS-{session}/.brainstorming/guidance-specification.md)
-   - Output: topic_framework_content
-
-2. **load_role_template**
-   - Action: Load {role-name} planning template
-   - Command: Read(~/.claude/workflows/cli-templates/planning-roles/{role}.md)
-   - Output: role_template_guidelines
-
-3. **load_session_metadata**
-   - Action: Load session metadata and original user intent
-   - Command: Read(.workflow/WFS-{session}/workflow-session.json)
-   - Output: session_context (contains original user prompt as PRIMARY reference)
+1. load_topic_framework → .workflow/active/WFS-{session}/.brainstorming/guidance-specification.md
+2. load_role_template → ~/.claude/workflows/cli-templates/planning-roles/{role}.md
+3. load_session_metadata → .workflow/active/WFS-{session}/workflow-session.json
+4. load_style_skill (ui-designer only, if style_skill_package) → .claude/skills/style-{style_skill_package}/
 
 ## Analysis Requirements
 **Primary Reference**: Original user prompt from workflow-session.json is authoritative
@@ -124,13 +154,9 @@ TOPIC: {user-provided-topic}
 **Template Integration**: Apply role template guidelines within framework structure
 
 ## Expected Deliverables
-1. **analysis.md**: Comprehensive {role-name} analysis addressing all framework discussion points
-   - **File Naming**: MUST start with `analysis` prefix (e.g., `analysis.md`, `analysis-1.md`, `analysis-2.md`)
-   - **FORBIDDEN**: Never use `recommendations.md` or any filename not starting with `analysis`
-   - **Auto-split if large**: If content >800 lines, split to `analysis-1.md`, `analysis-2.md` (max 3 files: analysis.md, analysis-1.md, analysis-2.md)
-   - **Content**: Includes both analysis AND recommendations sections within analysis files
-2. **Framework Reference**: Include @../guidance-specification.md reference in analysis
-3. **User Intent Alignment**: Validate analysis aligns with original user objectives from session_context
+1. **analysis.md** (optionally with analysis-{slug}.md sub-documents)
+2. **Framework Reference**: @../guidance-specification.md
+3. **User Intent Alignment**: Validate against session_context
 
 ## Completion Criteria
 - Address each discussion point from guidance-specification.md with {role-name} expertise
@@ -141,10 +167,11 @@ TOPIC: {user-provided-topic}
 "
 ```
 
-**Parallel Execution**:
+**Parallel Dispatch**:
 - Launch N agents simultaneously (one message with multiple Task calls)
+- Each agent task **attached** to orchestrator's TodoWrite
+- All agents execute concurrently, each attaching their own analysis sub-tasks
 - Each agent operates independently reading same guidance-specification.md
-- All agents update progress concurrently
 
 **Input**:
 - `selected_roles[]` from Phase 1
@@ -152,13 +179,40 @@ TOPIC: {user-provided-topic}
 - guidance-specification.md path
 
 **Validation**:
-- Each role creates `.workflow/WFS-{topic}/.brainstorming/{role}/analysis.md` (primary file)
-- If content is large (>800 lines), may split to `analysis-1.md`, `analysis-2.md` (max 3 files total)
-- **File naming pattern**: ALL files MUST start with `analysis` prefix (use `analysis*.md` for globbing)
-- **FORBIDDEN naming**: No `recommendations.md`, `recommendations-*.md`, or any non-`analysis` prefixed files
+- Each role creates `.workflow/active/WFS-{topic}/.brainstorming/{role}/analysis.md`
+- Optionally with `analysis-{slug}.md` sub-documents (max 5)
+- **File pattern**: `analysis*.md` for globbing
+- **FORBIDDEN**: `recommendations.md` or any non-`analysis` prefixed files
 - All N role analyses completed
 
-**TodoWrite**: Mark all N role agent tasks completed, phase 3 in_progress
+**TodoWrite Update (Phase 2 agents dispatched - tasks attached in parallel)**:
+```json
+[
+  {"content": "Phase 0: Parameter Parsing", "status": "completed", "activeForm": "Parsing count parameter"},
+  {"content": "Phase 1: Interactive Framework Generation", "status": "completed", "activeForm": "Executing artifacts interactive framework"},
+  {"content": "Phase 2: Parallel Role Analysis", "status": "in_progress", "activeForm": "Executing parallel role analysis"},
+  {"content": "  → Execute system-architect analysis", "status": "in_progress", "activeForm": "Executing system-architect analysis"},
+  {"content": "  → Execute ui-designer analysis", "status": "in_progress", "activeForm": "Executing ui-designer analysis"},
+  {"content": "  → Execute product-manager analysis", "status": "in_progress", "activeForm": "Executing product-manager analysis"},
+  {"content": "Phase 3: Synthesis Integration", "status": "pending", "activeForm": "Executing synthesis integration"}
+]
+```
+
+**Note**: Multiple Task dispatches **attach** N role analysis tasks simultaneously. Orchestrator **executes** these tasks in parallel.
+
+**Next Action**: Tasks attached → **Execute Phase 2.1-2.N** concurrently
+
+**TodoWrite Update (Phase 2 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Phase 0: Parameter Parsing", "status": "completed", "activeForm": "Parsing count parameter"},
+  {"content": "Phase 1: Interactive Framework Generation", "status": "completed", "activeForm": "Executing artifacts interactive framework"},
+  {"content": "Phase 2: Parallel Role Analysis", "status": "completed", "activeForm": "Executing parallel role analysis"},
+  {"content": "Phase 3: Synthesis Integration", "status": "pending", "activeForm": "Executing synthesis integration"}
+]
+```
+
+**Note**: Phase 2 parallel tasks completed and collapsed to summary.
 
 **After Phase 2**: Auto-continue to Phase 3 (synthesis)
 
@@ -166,7 +220,11 @@ TOPIC: {user-provided-topic}
 
 ### Phase 3: Synthesis Generation
 
-**Command**: `SlashCommand(command="/workflow:brainstorm:synthesis --session {sessionId}")`
+**Step 3: Dispatch** - Synthesis integration via synthesis command
+
+```javascript
+SlashCommand(command="/workflow:brainstorm:synthesis --session {sessionId}")
+```
 
 **What It Does**:
 - Load original user intent from workflow-session.json
@@ -177,16 +235,43 @@ TOPIC: {user-provided-topic}
 **Input**: `sessionId` from Phase 1
 
 **Validation**:
-- `.workflow/WFS-{topic}/.brainstorming/synthesis-specification.md` exists
+- `.workflow/active/WFS-{topic}/.brainstorming/synthesis-specification.md` exists
 - Synthesis references all role analyses
 
-**TodoWrite**: Mark phase 3 completed
+**TodoWrite Update (Phase 3 SlashCommand dispatched - tasks attached)**:
+```json
+[
+  {"content": "Phase 0: Parameter Parsing", "status": "completed", "activeForm": "Parsing count parameter"},
+  {"content": "Phase 1: Interactive Framework Generation", "status": "completed", "activeForm": "Executing artifacts interactive framework"},
+  {"content": "Phase 2: Parallel Role Analysis", "status": "completed", "activeForm": "Executing parallel role analysis"},
+  {"content": "Phase 3: Synthesis Integration", "status": "in_progress", "activeForm": "Executing synthesis integration"},
+  {"content": "  → Load role analysis files", "status": "in_progress", "activeForm": "Loading role analyses"},
+  {"content": "  → Integrate insights across roles", "status": "pending", "activeForm": "Integrating insights"},
+  {"content": "  → Generate synthesis specification", "status": "pending", "activeForm": "Generating synthesis"}
+]
+```
+
+**Note**: SlashCommand dispatch **attaches** synthesis' internal tasks. Orchestrator **executes** these tasks sequentially.
+
+**Next Action**: Tasks attached → **Execute Phase 3.1-3.3** sequentially
+
+**TodoWrite Update (Phase 3 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Phase 0: Parameter Parsing", "status": "completed", "activeForm": "Parsing count parameter"},
+  {"content": "Phase 1: Interactive Framework Generation", "status": "completed", "activeForm": "Executing artifacts interactive framework"},
+  {"content": "Phase 2: Parallel Role Analysis", "status": "completed", "activeForm": "Executing parallel role analysis"},
+  {"content": "Phase 3: Synthesis Integration", "status": "completed", "activeForm": "Executing synthesis integration"}
+]
+```
+
+**Note**: Phase 3 tasks completed and collapsed to summary.
 
 **Return to User**:
 ```
 Brainstorming complete for session: {sessionId}
 Roles analyzed: {count}
-Synthesis: .workflow/WFS-{topic}/.brainstorming/synthesis-specification.md
+Synthesis: .workflow/active/WFS-{topic}/.brainstorming/synthesis-specification.md
 
 ✅ Next Steps:
 1. /workflow:concept-clarify --session {sessionId}  # Optional refinement
@@ -195,49 +280,41 @@ Synthesis: .workflow/WFS-{topic}/.brainstorming/synthesis-specification.md
 
 ## TodoWrite Pattern
 
-```javascript
-// Initialize (before Phase 1)
-TodoWrite({todos: [
-  {"content": "Parse --count parameter from user input", "status": "in_progress", "activeForm": "Parsing count parameter"},
-  {"content": "Execute artifacts command for interactive framework generation", "status": "pending", "activeForm": "Executing artifacts interactive framework"},
-  {"content": "Load selected_roles from workflow-session.json", "status": "pending", "activeForm": "Loading selected roles"},
-  // Role agent tasks added dynamically after Phase 1 based on selected_roles count
-  {"content": "Execute synthesis command for final integration", "status": "pending", "activeForm": "Executing synthesis integration"}
-]})
+**Core Concept**: Dynamic task attachment and collapse for parallel brainstorming workflow with interactive framework generation and concurrent role analysis.
 
-// After Phase 1 (artifacts completes, roles loaded)
-// Note: artifacts EXTENDS this list by appending its Phase 1-5 sub-tasks
-TodoWrite({todos: [
-  {"content": "Parse --count parameter from user input", "status": "completed", "activeForm": "Parsing count parameter"},
-  {"content": "Execute artifacts command for interactive framework generation", "status": "completed", "activeForm": "Executing artifacts interactive framework"},
-  {"content": "Load selected_roles from workflow-session.json", "status": "in_progress", "activeForm": "Loading selected roles"},
-  {"content": "Execute system-architect analysis [conceptual-planning-agent]", "status": "pending", "activeForm": "Executing system-architect analysis"},
-  {"content": "Execute ui-designer analysis [conceptual-planning-agent]", "status": "pending", "activeForm": "Executing ui-designer analysis"},
-  {"content": "Execute product-manager analysis [conceptual-planning-agent]", "status": "pending", "activeForm": "Executing product-manager analysis"},
-  // ... (N role tasks based on --count parameter)
-  {"content": "Execute synthesis command for final integration", "status": "pending", "activeForm": "Executing synthesis integration"}
-]})
+### Key Principles
 
-// After Phase 2 (all agents launched in parallel)
-TodoWrite({todos: [
-  // ... previous completed tasks
-  {"content": "Load selected_roles from workflow-session.json", "status": "completed", "activeForm": "Loading selected roles"},
-  {"content": "Execute system-architect analysis [conceptual-planning-agent]", "status": "in_progress", "activeForm": "Executing system-architect analysis"},
-  {"content": "Execute ui-designer analysis [conceptual-planning-agent]", "status": "in_progress", "activeForm": "Executing ui-designer analysis"},
-  {"content": "Execute product-manager analysis [conceptual-planning-agent]", "status": "in_progress", "activeForm": "Executing product-manager analysis"},
-  // ... (all N agents in_progress simultaneously)
-  {"content": "Execute synthesis command for final integration", "status": "pending", "activeForm": "Executing synthesis integration"}
-]})
+1. **Task Attachment** (when SlashCommand/Task dispatched):
+   - Sub-command's or agent's internal tasks are **attached** to orchestrator's TodoWrite
+   - Phase 1: `/workflow:brainstorm:artifacts` attaches 5 internal tasks (Phase 1.1-1.5)
+   - Phase 2: Multiple `Task(conceptual-planning-agent)` calls attach N role analysis tasks simultaneously
+   - Phase 3: `/workflow:brainstorm:synthesis` attaches 3 internal tasks (Phase 3.1-3.3)
+   - First attached task marked as `in_progress`, others as `pending`
+   - Orchestrator **executes** these attached tasks (sequentially for Phase 1, 3; in parallel for Phase 2)
+
+2. **Task Collapse** (after sub-tasks complete):
+   - Remove detailed sub-tasks from TodoWrite
+   - **Collapse** to high-level phase summary
+   - Example: Phase 1.1-1.5 collapse to "Execute artifacts interactive framework generation: completed"
+   - Phase 2: Multiple role tasks collapse to "Execute parallel role analysis: completed"
+   - Phase 3: Synthesis tasks collapse to "Execute synthesis integration: completed"
+   - Maintains clean orchestrator-level view
+
+3. **Continuous Execution**:
+   - After collapse, automatically proceed to next pending phase
+   - No user intervention required between phases
+   - TodoWrite dynamically reflects current execution state
+
+**Lifecycle Summary**: Initial pending tasks → Phase 1 dispatched (artifacts tasks ATTACHED) → Artifacts sub-tasks executed → Phase 1 completed (tasks COLLAPSED) → Phase 2 dispatched (N role tasks ATTACHED in parallel) → Role analyses executed concurrently → Phase 2 completed (tasks COLLAPSED) → Phase 3 dispatched (synthesis tasks ATTACHED) → Synthesis sub-tasks executed → Phase 3 completed (tasks COLLAPSED) → Workflow complete.
+
+### Brainstorming Workflow Specific Features
+
+- **Phase 1**: Interactive framework generation with user Q&A (SlashCommand attachment)
+- **Phase 2**: Parallel role analysis execution with N concurrent agents (Task agent attachments)
+- **Phase 3**: Cross-role synthesis integration (SlashCommand attachment)
+- **Dynamic Role Count**: `--count N` parameter determines number of Phase 2 parallel tasks (default: 3, max: 9)
+- **Mixed Execution**: Sequential (Phase 1, 3) and Parallel (Phase 2) task execution
 
-// After Phase 2 (all agents complete)
-TodoWrite({todos: [
-  // ... previous completed tasks
-  {"content": "Execute system-architect analysis [conceptual-planning-agent]", "status": "completed", "activeForm": "Executing system-architect analysis"},
-  {"content": "Execute ui-designer analysis [conceptual-planning-agent]", "status": "completed", "activeForm": "Executing ui-designer analysis"},
-  {"content": "Execute product-manager analysis [conceptual-planning-agent]", "status": "completed", "activeForm": "Executing product-manager analysis"},
-  {"content": "Execute synthesis command for final integration", "status": "in_progress", "activeForm": "Executing synthesis integration"}
-]})
-```
 
 ## Input Processing
 
@@ -255,6 +332,34 @@ ELSE:
 EXECUTE: /workflow:brainstorm:artifacts "{topic}" --count {count_value}
 ```
 
+**Style-Skill Parameter Parsing**:
+```javascript
+// Extract --style-skill from user input
+IF user_input CONTAINS "--style-skill":
+    EXTRACT style_skill_name FROM "--style-skill package-name" pattern
+
+    // Validate SKILL package exists
+    skill_path = ".claude/skills/style-{style_skill_name}/SKILL.md"
+    IF file_exists(skill_path):
+        style_skill_package = style_skill_name
+        style_reference_path = ".workflow/reference_style/{style_skill_name}"
+        echo("✓ Style SKILL package found: style-{style_skill_name}")
+        echo("  Design reference: {style_reference_path}")
+    ELSE:
+        echo("⚠ WARNING: Style SKILL package not found: {style_skill_name}")
+        echo("  Expected location: {skill_path}")
+        echo("  Continuing without style reference...")
+        style_skill_package = null
+ELSE:
+    style_skill_package = null
+    echo("No style-skill specified, ui-designer will use default workflow")
+
+// Store for Phase 2 ui-designer context
+CONTEXT_VARS:
+    - style_skill_package: {style_skill_package}
+    - style_reference_path: {style_reference_path}
+```
+
 **Topic Structuring**:
 1. **Already Structured** → Pass directly to artifacts
    ```
@@ -270,32 +375,34 @@ EXECUTE: /workflow:brainstorm:artifacts "{topic}" --count {count_value}
 
 ## Session Management
 
-**⚡ FIRST ACTION**: Check for `.workflow/.active-*` markers before Phase 1
+**⚡ FIRST ACTION**: Check `.workflow/active/` for existing sessions before Phase 1
 
 **Multiple Sessions Support**:
-- Different Claude instances can have different active brainstorming sessions
-- If multiple active sessions found, prompt user to select
-- If single active session found, use it
-- If no active session exists, create `WFS-[topic-slug]`
+- Different Claude instances can have different brainstorming sessions
+- If multiple sessions found, prompt user to select
+- If single session found, use it
+- If no session exists, create `WFS-[topic-slug]`
 
 **Session Continuity**:
-- MUST use selected active session for all phases
+- MUST use selected session for all phases
 - Each role's context stored in session directory
 - Session isolation: Each session maintains independent state
 
 ## Output Structure
 
 **Phase 1 Output**:
-- `.workflow/WFS-{topic}/.brainstorming/guidance-specification.md` (framework content)
-- `.workflow/WFS-{topic}/workflow-session.json` (metadata: selected_roles[], topic, timestamps)
+- `.workflow/active/WFS-{topic}/.brainstorming/guidance-specification.md` (framework content)
+- `.workflow/active/WFS-{topic}/workflow-session.json` (metadata: selected_roles[], topic, timestamps, style_skill_package)
 
 **Phase 2 Output**:
-- `.workflow/WFS-{topic}/.brainstorming/{role}/analysis.md` (one per role)
+- `.workflow/active/WFS-{topic}/.brainstorming/{role}/analysis.md` (one per role)
+- `.superdesign/design_iterations/` (ui-designer artifacts, if --style-skill provided)
 
 **Phase 3 Output**:
-- `.workflow/WFS-{topic}/.brainstorming/synthesis-specification.md` (integrated analysis)
+- `.workflow/active/WFS-{topic}/.brainstorming/synthesis-specification.md` (integrated analysis)
 
 **⚠️ Storage Separation**: Guidance content in .md files, metadata in .json (no duplication)
+**⚠️ Style References**: When --style-skill provided, workflow-session.json stores style_skill_package name, ui-designer loads from `.claude/skills/style-{package-name}/`
 
 ## Available Roles
 
@@ -322,17 +429,13 @@ EXECUTE: /workflow:brainstorm:artifacts "{topic}" --count {count_value}
 
 **File Structure**:
 ```
-.workflow/WFS-[topic]/
-├── .active-brainstorming
+.workflow/active/WFS-[topic]/
 ├── workflow-session.json              # Session metadata ONLY
 └── .brainstorming/
     ├── guidance-specification.md      # Framework (Phase 1)
-    ├── {role-1}/
-    │   └── analysis.md                # Role analysis (Phase 2)
-    ├── {role-2}/
-    │   └── analysis.md
-    ├── {role-N}/
-    │   └── analysis.md
+    ├── {role}/
+    │   ├── analysis.md                # Main document (with optional @references)
+    │   └── analysis-{slug}.md         # Section documents (max 5)
     └── synthesis-specification.md     # Integration (Phase 3)
 ```
 

.claude/commands/workflow/brainstorm/data-architect.md

@@ -1,6 +1,6 @@
 ---
 name: data-architect
-description: Generate or update data-architect/analysis.md addressing guidance-specification discussion points
+description: Generate or update data-architect/analysis.md addressing guidance-specification discussion points for data architecture perspective
 argument-hint: "optional topic - uses existing framework if available"
 allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ---
@@ -47,10 +47,10 @@ allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ### Phase 1: Session & Framework Detection
 ```bash
 # Check active session and framework
-CHECK: .workflow/.active-* marker files
+CHECK: find .workflow/active/ -name "WFS-*" -type d
 IF active_session EXISTS:
     session_id = get_active_session()
-    brainstorm_dir = .workflow/WFS-{session}/.brainstorming/
+    brainstorm_dir = .workflow/active/WFS-{session}/.brainstorming/
 
     CHECK: brainstorm_dir/guidance-specification.md
     IF EXISTS:
@@ -87,13 +87,13 @@ Execute data-architect analysis for existing topic framework
 
 ## Context Loading
 ASSIGNED_ROLE: data-architect
-OUTPUT_LOCATION: .workflow/WFS-{session}/.brainstorming/data-architect/
+OUTPUT_LOCATION: .workflow/active/WFS-{session}/.brainstorming/data-architect/
 ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 ## Flow Control Steps
 1. **load_topic_framework**
    - Action: Load structured topic discussion framework
-   - Command: Read(.workflow/WFS-{session}/.brainstorming/guidance-specification.md)
+   - Command: Read(.workflow/active/WFS-{session}/.brainstorming/guidance-specification.md)
    - Output: topic_framework_content
 
 2. **load_role_template**
@@ -103,7 +103,7 @@ ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 3. **load_session_metadata**
    - Action: Load session metadata and existing context
-   - Command: Read(.workflow/WFS-{session}/workflow-session.json)
+   - Command: Read(.workflow/active/WFS-{session}/workflow-session.json)
    - Output: session_context
 
 ## Analysis Requirements
@@ -163,7 +163,7 @@ TodoWrite({
 
 ### Framework-Based Analysis
 ```
-.workflow/WFS-{session}/.brainstorming/data-architect/
+.workflow/active/WFS-{session}/.brainstorming/data-architect/
 └── analysis.md    # Structured analysis addressing guidance-specification.md discussion points
 ```
 
@@ -208,7 +208,7 @@ TodoWrite({
   "data_architect": {
     "status": "completed",
     "framework_addressed": true,
-    "output_location": ".workflow/WFS-{session}/.brainstorming/data-architect/analysis.md",
+    "output_location": ".workflow/active/WFS-{session}/.brainstorming/data-architect/analysis.md",
     "framework_reference": "@../guidance-specification.md"
   }
 }

.claude/commands/workflow/brainstorm/product-manager.md

@@ -1,6 +1,6 @@
 ---
 name: product-manager
-description: Generate or update product-manager/analysis.md addressing guidance-specification discussion points
+description: Generate or update product-manager/analysis.md addressing guidance-specification discussion points for product management perspective
 argument-hint: "optional topic - uses existing framework if available"
 allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ---
@@ -27,10 +27,10 @@ allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ### Phase 1: Session & Framework Detection
 ```bash
 # Check active session and framework
-CHECK: .workflow/.active-* marker files
+CHECK: find .workflow/active/ -name "WFS-*" -type d
 IF active_session EXISTS:
     session_id = get_active_session()
-    brainstorm_dir = .workflow/WFS-{session}/.brainstorming/
+    brainstorm_dir = .workflow/active/WFS-{session}/.brainstorming/
 
     CHECK: brainstorm_dir/guidance-specification.md
     IF EXISTS:
@@ -67,13 +67,13 @@ Execute product-manager analysis for existing topic framework
 
 ## Context Loading
 ASSIGNED_ROLE: product-manager
-OUTPUT_LOCATION: .workflow/WFS-{session}/.brainstorming/product-manager/
+OUTPUT_LOCATION: .workflow/active/WFS-{session}/.brainstorming/product-manager/
 ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 ## Flow Control Steps
 1. **load_topic_framework**
    - Action: Load structured topic discussion framework
-   - Command: Read(.workflow/WFS-{session}/.brainstorming/guidance-specification.md)
+   - Command: Read(.workflow/active/WFS-{session}/.brainstorming/guidance-specification.md)
    - Output: topic_framework_content
 
 2. **load_role_template**
@@ -83,7 +83,7 @@ ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 3. **load_session_metadata**
    - Action: Load session metadata and existing context
-   - Command: Read(.workflow/WFS-{session}/workflow-session.json)
+   - Command: Read(.workflow/active/WFS-{session}/workflow-session.json)
    - Output: session_context
 
 ## Analysis Requirements
@@ -143,7 +143,7 @@ TodoWrite({
 
 ### Framework-Based Analysis
 ```
-.workflow/WFS-{session}/.brainstorming/product-manager/
+.workflow/active/WFS-{session}/.brainstorming/product-manager/
 └── analysis.md    # Structured analysis addressing guidance-specification.md discussion points
 ```
 
@@ -188,7 +188,7 @@ TodoWrite({
   "product_manager": {
     "status": "completed",
     "framework_addressed": true,
-    "output_location": ".workflow/WFS-{session}/.brainstorming/product-manager/analysis.md",
+    "output_location": ".workflow/active/WFS-{session}/.brainstorming/product-manager/analysis.md",
     "framework_reference": "@../guidance-specification.md"
   }
 }

.claude/commands/workflow/brainstorm/product-owner.md

@@ -1,6 +1,6 @@
 ---
 name: product-owner
-description: Generate or update product-owner/analysis.md addressing guidance-specification discussion points
+description: Generate or update product-owner/analysis.md addressing guidance-specification discussion points for product ownership perspective
 argument-hint: "optional topic - uses existing framework if available"
 allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ---
@@ -27,10 +27,10 @@ allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ### Phase 1: Session & Framework Detection
 ```bash
 # Check active session and framework
-CHECK: .workflow/.active-* marker files
+CHECK: find .workflow/active/ -name "WFS-*" -type d
 IF active_session EXISTS:
     session_id = get_active_session()
-    brainstorm_dir = .workflow/WFS-{session}/.brainstorming/
+    brainstorm_dir = .workflow/active/WFS-{session}/.brainstorming/
 
     CHECK: brainstorm_dir/guidance-specification.md
     IF EXISTS:
@@ -67,13 +67,13 @@ Execute product-owner analysis for existing topic framework
 
 ## Context Loading
 ASSIGNED_ROLE: product-owner
-OUTPUT_LOCATION: .workflow/WFS-{session}/.brainstorming/product-owner/
+OUTPUT_LOCATION: .workflow/active/WFS-{session}/.brainstorming/product-owner/
 ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 ## Flow Control Steps
 1. **load_topic_framework**
    - Action: Load structured topic discussion framework
-   - Command: Read(.workflow/WFS-{session}/.brainstorming/guidance-specification.md)
+   - Command: Read(.workflow/active/WFS-{session}/.brainstorming/guidance-specification.md)
    - Output: topic_framework_content
 
 2. **load_role_template**
@@ -83,7 +83,7 @@ ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 3. **load_session_metadata**
    - Action: Load session metadata and existing context
-   - Command: Read(.workflow/WFS-{session}/workflow-session.json)
+   - Command: Read(.workflow/active/WFS-{session}/workflow-session.json)
    - Output: session_context
 
 ## Analysis Requirements
@@ -143,7 +143,7 @@ TodoWrite({
 
 ### Framework-Based Analysis
 ```
-.workflow/WFS-{session}/.brainstorming/product-owner/
+.workflow/active/WFS-{session}/.brainstorming/product-owner/
 └── analysis.md    # Structured analysis addressing guidance-specification.md discussion points
 ```
 
@@ -188,7 +188,7 @@ TodoWrite({
   "product_owner": {
     "status": "completed",
     "framework_addressed": true,
-    "output_location": ".workflow/WFS-{session}/.brainstorming/product-owner/analysis.md",
+    "output_location": ".workflow/active/WFS-{session}/.brainstorming/product-owner/analysis.md",
     "framework_reference": "@../guidance-specification.md"
   }
 }

.claude/commands/workflow/brainstorm/scrum-master.md

@@ -1,6 +1,6 @@
 ---
 name: scrum-master
-description: Generate or update scrum-master/analysis.md addressing guidance-specification discussion points
+description: Generate or update scrum-master/analysis.md addressing guidance-specification discussion points for Agile process perspective
 argument-hint: "optional topic - uses existing framework if available"
 allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ---
@@ -27,10 +27,10 @@ allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ### Phase 1: Session & Framework Detection
 ```bash
 # Check active session and framework
-CHECK: .workflow/.active-* marker files
+CHECK: find .workflow/active/ -name "WFS-*" -type d
 IF active_session EXISTS:
     session_id = get_active_session()
-    brainstorm_dir = .workflow/WFS-{session}/.brainstorming/
+    brainstorm_dir = .workflow/active/WFS-{session}/.brainstorming/
 
     CHECK: brainstorm_dir/guidance-specification.md
     IF EXISTS:
@@ -67,13 +67,13 @@ Execute scrum-master analysis for existing topic framework
 
 ## Context Loading
 ASSIGNED_ROLE: scrum-master
-OUTPUT_LOCATION: .workflow/WFS-{session}/.brainstorming/scrum-master/
+OUTPUT_LOCATION: .workflow/active/WFS-{session}/.brainstorming/scrum-master/
 ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 ## Flow Control Steps
 1. **load_topic_framework**
    - Action: Load structured topic discussion framework
-   - Command: Read(.workflow/WFS-{session}/.brainstorming/guidance-specification.md)
+   - Command: Read(.workflow/active/WFS-{session}/.brainstorming/guidance-specification.md)
    - Output: topic_framework_content
 
 2. **load_role_template**
@@ -83,7 +83,7 @@ ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 3. **load_session_metadata**
    - Action: Load session metadata and existing context
-   - Command: Read(.workflow/WFS-{session}/workflow-session.json)
+   - Command: Read(.workflow/active/WFS-{session}/workflow-session.json)
    - Output: session_context
 
 ## Analysis Requirements
@@ -143,7 +143,7 @@ TodoWrite({
 
 ### Framework-Based Analysis
 ```
-.workflow/WFS-{session}/.brainstorming/scrum-master/
+.workflow/active/WFS-{session}/.brainstorming/scrum-master/
 └── analysis.md    # Structured analysis addressing guidance-specification.md discussion points
 ```
 
@@ -188,7 +188,7 @@ TodoWrite({
   "scrum_master": {
     "status": "completed",
     "framework_addressed": true,
-    "output_location": ".workflow/WFS-{session}/.brainstorming/scrum-master/analysis.md",
+    "output_location": ".workflow/active/WFS-{session}/.brainstorming/scrum-master/analysis.md",
     "framework_reference": "@../guidance-specification.md"
   }
 }

.claude/commands/workflow/brainstorm/subject-matter-expert.md

@@ -1,6 +1,6 @@
 ---
 name: subject-matter-expert
-description: Generate or update subject-matter-expert/analysis.md addressing guidance-specification discussion points
+description: Generate or update subject-matter-expert/analysis.md addressing guidance-specification discussion points for domain expertise perspective
 argument-hint: "optional topic - uses existing framework if available"
 allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ---
@@ -27,10 +27,10 @@ allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ### Phase 1: Session & Framework Detection
 ```bash
 # Check active session and framework
-CHECK: .workflow/.active-* marker files
+CHECK: find .workflow/active/ -name "WFS-*" -type d
 IF active_session EXISTS:
     session_id = get_active_session()
-    brainstorm_dir = .workflow/WFS-{session}/.brainstorming/
+    brainstorm_dir = .workflow/active/WFS-{session}/.brainstorming/
 
     CHECK: brainstorm_dir/guidance-specification.md
     IF EXISTS:
@@ -67,13 +67,13 @@ Execute subject-matter-expert analysis for existing topic framework
 
 ## Context Loading
 ASSIGNED_ROLE: subject-matter-expert
-OUTPUT_LOCATION: .workflow/WFS-{session}/.brainstorming/subject-matter-expert/
+OUTPUT_LOCATION: .workflow/active/WFS-{session}/.brainstorming/subject-matter-expert/
 ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 ## Flow Control Steps
 1. **load_topic_framework**
    - Action: Load structured topic discussion framework
-   - Command: Read(.workflow/WFS-{session}/.brainstorming/guidance-specification.md)
+   - Command: Read(.workflow/active/WFS-{session}/.brainstorming/guidance-specification.md)
    - Output: topic_framework_content
 
 2. **load_role_template**
@@ -83,7 +83,7 @@ ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 3. **load_session_metadata**
    - Action: Load session metadata and existing context
-   - Command: Read(.workflow/WFS-{session}/workflow-session.json)
+   - Command: Read(.workflow/active/WFS-{session}/workflow-session.json)
    - Output: session_context
 
 ## Analysis Requirements
@@ -143,7 +143,7 @@ TodoWrite({
 
 ### Framework-Based Analysis
 ```
-.workflow/WFS-{session}/.brainstorming/subject-matter-expert/
+.workflow/active/WFS-{session}/.brainstorming/subject-matter-expert/
 └── analysis.md    # Structured analysis addressing guidance-specification.md discussion points
 ```
 
@@ -188,7 +188,7 @@ TodoWrite({
   "subject_matter_expert": {
     "status": "completed",
     "framework_addressed": true,
-    "output_location": ".workflow/WFS-{session}/.brainstorming/subject-matter-expert/analysis.md",
+    "output_location": ".workflow/active/WFS-{session}/.brainstorming/subject-matter-expert/analysis.md",
     "framework_reference": "@../guidance-specification.md"
   }
 }

.claude/commands/workflow/brainstorm/synthesis.md

@@ -1,326 +1,319 @@
 ---
 name: synthesis
-description: Clarify and refine role analyses through intelligent Q&A and targeted updates
+description: Clarify and refine role analyses through intelligent Q&A and targeted updates with synthesis agent
 argument-hint: "[optional: --session session-id]"
-allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*), Edit(*), Glob(*)
+allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*), Edit(*), Glob(*), AskUserQuestion(*)
 ---
 
 ## Overview
 
-Three-phase workflow to eliminate ambiguities and enhance conceptual depth in role analyses:
+Six-phase workflow to eliminate ambiguities and enhance conceptual depth in role analyses:
 
-**Phase 1-2 (Main Flow)**: Session detection → File discovery → Path preparation
+**Phase 1-2**: Session detection → File discovery → Path preparation
+**Phase 3A**: Cross-role analysis agent → Generate recommendations
+**Phase 4**: User selects enhancements → User answers clarifications (via AskUserQuestion)
+**Phase 5**: Parallel update agents (one per role)
+**Phase 6**: Context package update → Metadata update → Completion report
 
-**Phase 3A (Analysis Agent)**: Cross-role analysis → Generate recommendations
-
-**Phase 4 (Main Flow)**: User selects enhancements → User answers clarifications → Build update plan
-
-**Phase 5 (Parallel Update Agents)**: Each agent updates ONE role document → Parallel execution
-
-**Phase 6 (Main Flow)**: Metadata update → Completion report
-
-**Key Features**:
-- Multi-agent architecture (analysis agent + parallel update agents)
-- Clear separation: Agent analysis vs Main flow interaction
-- Parallel document updates (one agent per role)
-- User intent alignment validation
+All user interactions use AskUserQuestion tool (max 4 questions per call, multi-round).
 
 **Document Flow**:
 - Input: `[role]/analysis*.md`, `guidance-specification.md`, session metadata
 - Output: Updated `[role]/analysis*.md` with Enhancements + Clarifications sections
 
+---
+
+## Quick Reference
+
+### Phase Summary
+
+| Phase | Goal | Executor | Output |
+|-------|------|----------|--------|
+| 1 | Session detection | Main flow | session_id, brainstorm_dir |
+| 2 | File discovery | Main flow | role_analysis_paths |
+| 3A | Cross-role analysis | Agent | enhancement_recommendations |
+| 4 | User interaction | Main flow + AskUserQuestion | update_plan |
+| 5 | Document updates | Parallel agents | Updated analysis*.md |
+| 6 | Finalization | Main flow | context-package.json, report |
+
+### AskUserQuestion Pattern
+
+```javascript
+// Enhancement selection (multi-select)
+AskUserQuestion({
+  questions: [{
+    question: "请选择要应用的改进建议",
+    header: "改进选择",
+    multiSelect: true,
+    options: [
+      { label: "EP-001: API Contract", description: "添加详细的请求/响应 schema 定义" },
+      { label: "EP-002: User Intent", description: "明确用户需求优先级和验收标准" }
+    ]
+  }]
+})
+
+// Clarification questions (single-select, multi-round)
+AskUserQuestion({
+  questions: [
+    {
+      question: "MVP 阶段的核心目标是什么？",
+      header: "用户意图",
+      multiSelect: false,
+      options: [
+        { label: "快速验证", description: "最小功能集，快速上线获取反馈" },
+        { label: "技术壁垒", description: "完善架构，为长期发展打基础" },
+        { label: "功能完整", description: "覆盖所有规划功能，延迟上线" }
+      ]
+    }
+  ]
+})
+```
+
+---
+
 ## Task Tracking
 
 ```json
 [
-  {"content": "Detect session and validate analyses", "status": "in_progress", "activeForm": "Detecting session"},
+  {"content": "Detect session and validate analyses", "status": "pending", "activeForm": "Detecting session"},
   {"content": "Discover role analysis file paths", "status": "pending", "activeForm": "Discovering paths"},
-  {"content": "Execute analysis agent (cross-role analysis)", "status": "pending", "activeForm": "Executing analysis agent"},
-  {"content": "Present enhancements for user selection", "status": "pending", "activeForm": "Presenting enhancements"},
-  {"content": "Generate and present clarification questions", "status": "pending", "activeForm": "Clarifying with user"},
-  {"content": "Build update plan from user input", "status": "pending", "activeForm": "Building update plan"},
-  {"content": "Execute parallel update agents (one per role)", "status": "pending", "activeForm": "Updating documents in parallel"},
-  {"content": "Update session metadata and generate report", "status": "pending", "activeForm": "Finalizing session"}
+  {"content": "Execute analysis agent (cross-role analysis)", "status": "pending", "activeForm": "Executing analysis"},
+  {"content": "Present enhancements via AskUserQuestion", "status": "pending", "activeForm": "Selecting enhancements"},
+  {"content": "Clarification questions via AskUserQuestion", "status": "pending", "activeForm": "Clarifying"},
+  {"content": "Execute parallel update agents", "status": "pending", "activeForm": "Updating documents"},
+  {"content": "Update context package and metadata", "status": "pending", "activeForm": "Finalizing"}
 ]
 ```
 
+---
+
 ## Execution Phases
 
 ### Phase 1: Discovery & Validation
 
-1. **Detect Session**: Use `--session` parameter or `.workflow/.active-*` marker
+1. **Detect Session**: Use `--session` parameter or find `.workflow/active/WFS-*`
 2. **Validate Files**:
    - `guidance-specification.md` (optional, warn if missing)
    - `*/analysis*.md` (required, error if empty)
-3. **Load User Intent**: Extract from `workflow-session.json` (project/description field)
+3. **Load User Intent**: Extract from `workflow-session.json`
 
 ### Phase 2: Role Discovery & Path Preparation
 
 **Main flow prepares file paths for Agent**:
 
 1. **Discover Analysis Files**:
-   - Glob(.workflow/WFS-{session}/.brainstorming/*/analysis*.md)
-   - Supports: analysis.md, analysis-1.md, analysis-2.md, analysis-3.md
-   - Validate: At least one file exists (error if empty)
+   - Glob: `.workflow/active/WFS-{session}/.brainstorming/*/analysis*.md`
+   - Supports: analysis.md + analysis-{slug}.md (max 5)
 
 2. **Extract Role Information**:
-   - `role_analysis_paths`: Relative paths from brainstorm_dir
-   - `participating_roles`: Role names extracted from directory paths
+   - `role_analysis_paths`: Relative paths
+   - `participating_roles`: Role names from directories
 
-3. **Pass to Agent** (Phase 3):
-   - `session_id`
-   - `brainstorm_dir`: .workflow/WFS-{session}/.brainstorming/
-   - `role_analysis_paths`: ["product-manager/analysis.md", "system-architect/analysis-1.md", ...]
-   - `participating_roles`: ["product-manager", "system-architect", ...]
-
-**Main Flow Responsibility**: File discovery and path preparation only (NO file content reading)
+3. **Pass to Agent**: session_id, brainstorm_dir, role_analysis_paths, participating_roles
 
 ### Phase 3A: Analysis & Enhancement Agent
 
-**First agent call**: Cross-role analysis and generate enhancement recommendations
+**Agent executes cross-role analysis**:
 
-```bash
-Task(conceptual-planning-agent): "
+```javascript
+Task(conceptual-planning-agent, `
 ## Agent Mission
-Analyze role documents, identify conflicts/gaps, and generate enhancement recommendations
+Analyze role documents, identify conflicts/gaps, generate enhancement recommendations
 
-## Input from Main Flow
-- brainstorm_dir: {brainstorm_dir}
-- role_analysis_paths: {role_analysis_paths}
-- participating_roles: {participating_roles}
+## Input
+- brainstorm_dir: ${brainstorm_dir}
+- role_analysis_paths: ${role_analysis_paths}
+- participating_roles: ${participating_roles}
 
-## Execution Instructions
-[FLOW_CONTROL]
+## Flow Control Steps
+1. load_session_metadata → Read workflow-session.json
+2. load_role_analyses → Read all analysis files
+3. cross_role_analysis → Identify consensus, conflicts, gaps, ambiguities
+4. generate_recommendations → Format as EP-001, EP-002, ...
 
-### Flow Control Steps
-**AGENT RESPONSIBILITY**: Execute these analysis steps sequentially with context accumulation:
-
-1. **load_session_metadata**
-   - Action: Load original user intent as primary reference
-   - Command: Read({brainstorm_dir}/../workflow-session.json)
-   - Output: original_user_intent (from project/description field)
-
-2. **load_role_analyses**
-   - Action: Load all role analysis documents
-   - Command: For each path in role_analysis_paths: Read({brainstorm_dir}/{path})
-   - Output: role_analyses_content_map = {role_name: content}
-
-3. **cross_role_analysis**
-   - Action: Identify consensus themes, conflicts, gaps, underspecified areas
-   - Output: consensus_themes, conflicting_views, gaps_list, ambiguities
-
-4. **generate_recommendations**
-   - Action: Convert cross-role analysis findings into structured enhancement recommendations
-   - Format: EP-001, EP-002, ... (sequential numbering)
-   - Fields: id, title, affected_roles, category, current_state, enhancement, rationale, priority
-   - Taxonomy: Map to 9 categories (User Intent, Requirements, Architecture, UX, Feasibility, Risk, Process, Decisions, Terminology)
-   - Output: enhancement_recommendations (JSON array)
-
-### Output to Main Flow
-Return JSON array:
+## Output Format
 [
   {
-    \"id\": \"EP-001\",
-    \"title\": \"API Contract Specification\",
-    \"affected_roles\": [\"system-architect\", \"api-designer\"],
-    \"category\": \"Architecture\",
-    \"current_state\": \"High-level API descriptions\",
-    \"enhancement\": \"Add detailed contract definitions with request/response schemas\",
-    \"rationale\": \"Enables precise implementation and testing\",
-    \"priority\": \"High\"
-  },
-  ...
+    "id": "EP-001",
+    "title": "API Contract Specification",
+    "affected_roles": ["system-architect", "api-designer"],
+    "category": "Architecture",
+    "current_state": "High-level API descriptions",
+    "enhancement": "Add detailed contract definitions",
+    "rationale": "Enables precise implementation",
+    "priority": "High"
+  }
 ]
-
-"
+`)
 ```
 
-### Phase 4: Main Flow User Interaction
+### Phase 4: User Interaction
 
-**Main flow handles all user interaction via text output**:
+**All interactions via AskUserQuestion (Chinese questions)**
 
-**⚠️ CRITICAL**: ALL questions MUST use Chinese (所有问题必须用中文) for better user understanding
+#### Step 1: Enhancement Selection
 
-1. **Present Enhancement Options** (multi-select):
-```markdown
-===== Enhancement 选择 =====
+```javascript
+// If enhancements > 4, split into multiple rounds
+const enhancements = [...]; // from Phase 3A
+const BATCH_SIZE = 4;
 
-请选择要应用的改进建议（可多选）：
+for (let i = 0; i < enhancements.length; i += BATCH_SIZE) {
+  const batch = enhancements.slice(i, i + BATCH_SIZE);
 
-a) EP-001: API Contract Specification
-   影响角色：system-architect, api-designer
-   说明：添加详细的请求/响应 schema 定义
+  AskUserQuestion({
+    questions: [{
+      question: `请选择要应用的改进建议 (第${Math.floor(i/BATCH_SIZE)+1}轮)`,
+      header: "改进选择",
+      multiSelect: true,
+      options: batch.map(ep => ({
+        label: `${ep.id}: ${ep.title}`,
+        description: `影响: ${ep.affected_roles.join(', ')} | ${ep.enhancement}`
+      }))
+    }]
+  })
 
-b) EP-002: User Intent Validation
-   影响角色：product-manager, ux-expert
-   说明：明确用户需求优先级和验收标准
+  // Store selections before next round
+}
 
-c) EP-003: Error Handling Strategy
-   影响角色：system-architect
-   说明：统一异常处理和降级方案
-
-支持格式：1abc 或 1a 1b 1c 或 1a,b,c
-请输入选择（可跳过输入 skip）：
+// User can also skip: provide "跳过" option
 ```
 
-2. **Generate Clarification Questions** (based on analysis agent output):
-   - ✅ **ALL questions in Chinese (所有问题必须用中文)**
-   - Use 9-category taxonomy scan results
-   - Prioritize most critical questions (no hard limit)
-   - Each with 2-4 options + descriptions
+#### Step 2: Clarification Questions
 
-3. **Interactive Clarification Loop** (max 10 questions per round):
-```markdown
-===== Clarification 问题 (第 1/2 轮) =====
+```javascript
+// Generate questions based on 9-category taxonomy scan
+// Categories: User Intent, Requirements, Architecture, UX, Feasibility, Risk, Process, Decisions, Terminology
 
-【问题1 - 用户意图】MVP 阶段的核心目标是什么？
-a) 快速验证市场需求
-   说明：最小功能集，快速上线获取反馈
-b) 建立技术壁垒
-   说明：完善架构，为长期发展打基础
-c) 实现功能完整性
-   说明：覆盖所有规划功能，延迟上线
+const clarifications = [...]; // from analysis
+const BATCH_SIZE = 4;
 
-【问题2 - 架构决策】技术栈选择的优先考虑因素？
-a) 团队熟悉度
-   说明：使用现有技术栈，降低学习成本
-b) 技术先进性
-   说明：采用新技术，提升竞争力
-c) 生态成熟度
-   说明：选择成熟方案，保证稳定性
+for (let i = 0; i < clarifications.length; i += BATCH_SIZE) {
+  const batch = clarifications.slice(i, i + BATCH_SIZE);
+  const currentRound = Math.floor(i / BATCH_SIZE) + 1;
+  const totalRounds = Math.ceil(clarifications.length / BATCH_SIZE);
 
-...（最多10个问题）
+  AskUserQuestion({
+    questions: batch.map(q => ({
+      question: q.question,
+      header: q.category.substring(0, 12),
+      multiSelect: false,
+      options: q.options.map(opt => ({
+        label: opt.label,
+        description: opt.description
+      }))
+    }))
+  })
 
-请回答 (格式: 1a 2b 3c...)：
+  // Store answers before next round
+}
 ```
 
-Wait for user input → Parse all answers in batch → Continue to next round if needed
+### Question Guidelines
 
-4. **Build Update Plan**:
-```
+**Target**: 开发者（理解技术但需要从用户需求出发）
+
+**Question Structure**: `[跨角色分析发现] + [需要澄清的决策点]`
+**Option Structure**: `标签：[具体方案] + 说明：[业务影响] + [技术权衡]`
+
+**9-Category Taxonomy**:
+
+| Category | Focus | Example Question Pattern |
+|----------|-------|--------------------------|
+| User Intent | 用户目标 | "MVP阶段核心目标？" + 验证/壁垒/完整性 |
+| Requirements | 需求细化 | "功能优先级如何排序？" + 核心/增强/可选 |
+| Architecture | 架构决策 | "技术栈选择考量？" + 熟悉度/先进性/成熟度 |
+| UX | 用户体验 | "交互复杂度取舍？" + 简洁/丰富/渐进 |
+| Feasibility | 可行性 | "资源约束下的范围？" + 最小/标准/完整 |
+| Risk | 风险管理 | "风险容忍度？" + 保守/平衡/激进 |
+| Process | 流程规范 | "迭代节奏？" + 快速/稳定/灵活 |
+| Decisions | 决策确认 | "冲突解决方案？" + 方案A/方案B/折中 |
+| Terminology | 术语统一 | "统一使用哪个术语？" + 术语A/术语B |
+
+**Quality Rules**:
+
+**MUST Include**:
+- ✅ All questions in Chinese (用中文提问)
+- ✅ 基于跨角色分析的具体发现
+- ✅ 选项包含业务影响说明
+- ✅ 解决实际的模糊点或冲突
+
+**MUST Avoid**:
+- ❌ 与角色分析无关的通用问题
+- ❌ 重复已在 artifacts 阶段确认的内容
+- ❌ 过于细节的实现级问题
+
+#### Step 3: Build Update Plan
+
+```javascript
 update_plan = {
   "role1": {
-    "enhancements": [EP-001, EP-003],
+    "enhancements": ["EP-001", "EP-003"],
     "clarifications": [
-      {"question": "...", "answer": "...", "category": "..."},
-      ...
+      {"question": "...", "answer": "...", "category": "..."}
     ]
   },
   "role2": {
-    "enhancements": [EP-002],
+    "enhancements": ["EP-002"],
     "clarifications": [...]
-  },
-  ...
+  }
 }
 ```
 
 ### Phase 5: Parallel Document Update Agents
 
-**Parallel agent calls** (one per role needing updates):
+**Execute in parallel** (one agent per role):
 
-```bash
-# Execute in parallel using single message with multiple Task calls
-
-Task(conceptual-planning-agent): "
+```javascript
+// Single message with multiple Task calls for parallelism
+Task(conceptual-planning-agent, `
 ## Agent Mission
-Apply user-confirmed enhancements and clarifications to {role1} analysis document
+Apply enhancements and clarifications to ${role} analysis
 
-## Agent Intent
-- **Goal**: Integrate synthesis results into role-specific analysis
-- **Scope**: Update ONLY {role1}/analysis.md (isolated, no cross-role dependencies)
-- **Constraints**: Preserve original insights, add refinements without deletion
+## Input
+- role: ${role}
+- analysis_path: ${brainstorm_dir}/${role}/analysis.md
+- enhancements: ${role_enhancements}
+- clarifications: ${role_clarifications}
+- original_user_intent: ${intent}
 
-## Input from Main Flow
-- role: {role1}
-- analysis_path: {brainstorm_dir}/{role1}/analysis.md
-- enhancements: [EP-001, EP-003] (user-selected improvements)
-- clarifications: [{question, answer, category}, ...] (user-confirmed answers)
-- original_user_intent: {from session metadata}
+## Flow Control Steps
+1. load_current_analysis → Read analysis file
+2. add_clarifications_section → Insert Q&A section
+3. apply_enhancements → Integrate into relevant sections
+4. resolve_contradictions → Remove conflicts
+5. enforce_terminology → Align terminology
+6. validate_intent → Verify alignment with user intent
+7. write_updated_file → Save changes
 
-## Execution Instructions
-[FLOW_CONTROL]
-
-### Flow Control Steps
-**AGENT RESPONSIBILITY**: Execute these update steps sequentially:
-
-1. **load_current_analysis**
-   - Action: Load existing role analysis document
-   - Command: Read({brainstorm_dir}/{role1}/analysis.md)
-   - Output: current_analysis_content
-
-2. **add_clarifications_section**
-   - Action: Insert Clarifications section with Q&A
-   - Format: \"## Clarifications\\n### Session {date}\\n- **Q**: {question} (Category: {category})\\n  **A**: {answer}\"
-   - Output: analysis_with_clarifications
-
-3. **apply_enhancements**
-   - Action: Integrate EP-001, EP-003 into relevant sections
-   - Strategy: Locate section by category (Architecture → Architecture section, UX → User Experience section)
-   - Output: analysis_with_enhancements
-
-4. **resolve_contradictions**
-   - Action: Remove conflicts between original content and clarifications/enhancements
-   - Output: contradiction_free_analysis
-
-5. **enforce_terminology_consistency**
-   - Action: Align all terminology with user-confirmed choices from clarifications
-   - Output: terminology_consistent_analysis
-
-6. **validate_user_intent_alignment**
-   - Action: Verify all updates support original_user_intent
-   - Output: validated_analysis
-
-7. **write_updated_file**
-   - Action: Save final analysis document
-   - Command: Write({brainstorm_dir}/{role1}/analysis.md, validated_analysis)
-   - Output: File update confirmation
-
-### Output
-Updated {role1}/analysis.md with Clarifications section + enhanced content
-")
-
-Task(conceptual-planning-agent): "
-## Agent Mission
-Apply user-confirmed enhancements and clarifications to {role2} analysis document
-
-## Agent Intent
-- **Goal**: Integrate synthesis results into role-specific analysis
-- **Scope**: Update ONLY {role2}/analysis.md (isolated, no cross-role dependencies)
-- **Constraints**: Preserve original insights, add refinements without deletion
-
-## Input from Main Flow
-- role: {role2}
-- analysis_path: {brainstorm_dir}/{role2}/analysis.md
-- enhancements: [EP-002] (user-selected improvements)
-- clarifications: [{question, answer, category}, ...] (user-confirmed answers)
-- original_user_intent: {from session metadata}
-
-## Execution Instructions
-[FLOW_CONTROL]
-
-### Flow Control Steps
-**AGENT RESPONSIBILITY**: Execute same 7 update steps as {role1} agent (load → clarifications → enhancements → contradictions → terminology → validation → write)
-
-### Output
-Updated {role2}/analysis.md with Clarifications section + enhanced content
-")
-
-# ... repeat for each role in update_plan
+## Output
+Updated ${role}/analysis.md
+`)
 ```
 
 **Agent Characteristics**:
-- **Intent**: Integrate user-confirmed synthesis results (NOT generate new analysis)
-- **Isolation**: Each agent updates exactly ONE role (parallel execution safe)
-- **Context**: Minimal - receives only role-specific enhancements + clarifications
-- **Dependencies**: Zero cross-agent dependencies (full parallelism)
+- **Isolation**: Each agent updates exactly ONE role (parallel safe)
+- **Dependencies**: Zero cross-agent dependencies
 - **Validation**: All updates must align with original_user_intent
 
-### Phase 6: Completion & Metadata Update
+### Phase 6: Finalization
 
-**Main flow finalizes**:
+#### Step 1: Update Context Package
+
+```javascript
+// Sync updated analyses to context-package.json
+const context_pkg = Read(".workflow/active/WFS-{session}/.process/context-package.json")
+
+// Update guidance-specification if exists
+// Update synthesis-specification if exists
+// Re-read all role analysis files
+// Update metadata timestamps
+
+Write(context_pkg_path, JSON.stringify(context_pkg))
+```
+
+#### Step 2: Update Session Metadata
 
-1. Wait for all parallel agents to complete
-2. Update workflow-session.json:
 ```json
 {
   "phases": {
@@ -330,15 +323,13 @@ Updated {role2}/analysis.md with Clarifications section + enhanced content
       "completed_at": "timestamp",
       "participating_roles": [...],
       "clarification_results": {
-        "enhancements_applied": ["EP-001", "EP-002", ...],
+        "enhancements_applied": ["EP-001", "EP-002"],
         "questions_asked": 3,
-        "categories_clarified": ["Architecture", "UX", ...],
-        "roles_updated": ["role1", "role2", ...],
-        "outstanding_items": []
+        "categories_clarified": ["Architecture", "UX"],
+        "roles_updated": ["role1", "role2"]
       },
       "quality_metrics": {
         "user_intent_alignment": "validated",
-        "requirement_coverage": "comprehensive",
         "ambiguity_resolution": "complete",
         "terminology_consistency": "enforced"
       }
@@ -347,7 +338,8 @@ Updated {role2}/analysis.md with Clarifications section + enhanced content
 }
 ```
 
-3. Generate completion report (show to user):
+#### Step 3: Completion Report
+
 ```markdown
 ## ✅ Clarification Complete
 
@@ -359,9 +351,11 @@ Updated {role2}/analysis.md with Clarifications section + enhanced content
 ✅ PROCEED: `/workflow:plan --session WFS-{session-id}`
 ```
 
+---
+
 ## Output
 
-**Location**: `.workflow/WFS-{session}/.brainstorming/[role]/analysis*.md` (in-place updates)
+**Location**: `.workflow/active/WFS-{session}/.brainstorming/[role]/analysis*.md`
 
 **Updated Structure**:
 ```markdown
@@ -381,58 +375,24 @@ Updated {role2}/analysis.md with Clarifications section + enhanced content
 - Ambiguities resolved, placeholders removed
 - Consistent terminology
 
-## Session Metadata
-
-Update `workflow-session.json`:
-
-```json
-{
-  "phases": {
-    "BRAINSTORM": {
-      "status": "clarification_completed",
-      "clarification_completed": true,
-      "completed_at": "timestamp",
-      "participating_roles": ["product-manager", "system-architect", ...],
-      "clarification_results": {
-        "questions_asked": 3,
-        "categories_clarified": ["Architecture & Design", ...],
-        "roles_updated": ["system-architect", "ui-designer", ...],
-        "outstanding_items": []
-      },
-      "quality_metrics": {
-        "user_intent_alignment": "validated",
-        "requirement_coverage": "comprehensive",
-        "ambiguity_resolution": "complete",
-        "terminology_consistency": "enforced",
-        "decision_transparency": "documented"
-      }
-    }
-  }
-}
-```
+---
 
 ## Quality Checklist
 
 **Content**:
-- All role analyses loaded/analyzed
-- Cross-role analysis (consensus, conflicts, gaps)
-- 9-category ambiguity scan
-- Questions prioritized
-- Clarifications documented
+- ✅ All role analyses loaded/analyzed
+- ✅ Cross-role analysis (consensus, conflicts, gaps)
+- ✅ 9-category ambiguity scan
+- ✅ Questions prioritized
 
 **Analysis**:
-- User intent validated
-- Cross-role synthesis complete
-- Ambiguities resolved
-- Correct roles updated
-- Terminology consistent
-- Contradictions removed
+- ✅ User intent validated
+- ✅ Cross-role synthesis complete
+- ✅ Ambiguities resolved
+- ✅ Terminology consistent
 
 **Documents**:
-- Clarifications section formatted
-- Sections reflect answers
-- No placeholders (TODO/TBD)
-- Valid Markdown
-- Cross-references maintained
-
-
+- ✅ Clarifications section formatted
+- ✅ Sections reflect answers
+- ✅ No placeholders (TODO/TBD)
+- ✅ Valid Markdown

.claude/commands/workflow/brainstorm/system-architect.md

@@ -1,6 +1,6 @@
 ---
 name: system-architect
-description: Generate or update system-architect/analysis.md addressing guidance-specification discussion points
+description: Generate or update system-architect/analysis.md addressing guidance-specification discussion points for system architecture perspective
 argument-hint: "optional topic - uses existing framework if available"
 allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ---
@@ -46,10 +46,10 @@ allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ### Phase 1: Session & Framework Detection
 ```bash
 # Check active session and framework
-CHECK: .workflow/.active-* marker files
+CHECK: find .workflow/active/ -name "WFS-*" -type d
 IF active_session EXISTS:
     session_id = get_active_session()
-    brainstorm_dir = .workflow/WFS-{session}/.brainstorming/
+    brainstorm_dir = .workflow/active/WFS-{session}/.brainstorming/
 
     CHECK: brainstorm_dir/guidance-specification.md
     IF EXISTS:
@@ -81,6 +81,7 @@ ELSE:
 **Framework-Based Analysis** (when guidance-specification.md exists):
 ```bash
 Task(subagent_type="conceptual-planning-agent",
+     run_in_background=false,
      prompt="Generate system architect analysis addressing topic framework
 
      ## Framework Integration Required
@@ -136,6 +137,7 @@ Task(subagent_type="conceptual-planning-agent",
 # For existing analysis updates
 IF update_mode = "incremental":
     Task(subagent_type="conceptual-planning-agent",
+         run_in_background=false,
          prompt="Update existing system architect analysis
 
          ## Current Analysis Context
@@ -162,7 +164,7 @@ IF update_mode = "incremental":
 
 ### Output Files
 ```
-.workflow/WFS-[topic]/.brainstorming/
+.workflow/active/WFS-[topic]/.brainstorming/
 ├── guidance-specification.md          # Input: Framework (if exists)
 └── system-architect/
     └── analysis.md            # ★ OUTPUT: Framework-based analysis
@@ -186,8 +188,8 @@ IF update_mode = "incremental":
 ### ⚠️ Session Management - FIRST STEP
 Session detection and selection:
 ```bash
-# Check for active sessions
-active_sessions=$(find .workflow -name ".active-*" 2>/dev/null)
+# Check for existing sessions
+existing_sessions=$(find .workflow/active/ -name "WFS-*" -type d 2>/dev/null)
 if [ multiple_sessions ]; then
   prompt_user_to_select_session()
 else
@@ -279,7 +281,7 @@ TodoWrite tracking for two-step process:
 
 ### Output Location
 ```
-.workflow/WFS-{topic-slug}/.brainstorming/system-architect/
+.workflow/active/WFS-{topic-slug}/.brainstorming/system-architect/
 ├── analysis.md                 # Primary architecture analysis
 ├── architecture-design.md      # Detailed system design and diagrams
 ├── technology-stack.md         # Technology stack recommendations and justifications
@@ -340,7 +342,7 @@ Upon completion, update `workflow-session.json`:
       "system_architect": {
         "status": "completed",
         "completed_at": "timestamp",
-        "output_directory": ".workflow/WFS-{topic}/.brainstorming/system-architect/",
+        "output_directory": ".workflow/active/WFS-{topic}/.brainstorming/system-architect/",
         "key_insights": ["scalability_bottleneck", "architecture_pattern", "technology_recommendation"]
       }
     }

.claude/commands/workflow/brainstorm/ui-designer.md

@@ -1,6 +1,6 @@
 ---
 name: ui-designer
-description: Generate or update ui-designer/analysis.md addressing guidance-specification discussion points
+description: Generate or update ui-designer/analysis.md addressing guidance-specification discussion points for UI design perspective
 argument-hint: "optional topic - uses existing framework if available"
 allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ---
@@ -48,10 +48,10 @@ allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ### Phase 1: Session & Framework Detection
 ```bash
 # Check active session and framework
-CHECK: .workflow/.active-* marker files
+CHECK: find .workflow/active/ -name "WFS-*" -type d
 IF active_session EXISTS:
     session_id = get_active_session()
-    brainstorm_dir = .workflow/WFS-{session}/.brainstorming/
+    brainstorm_dir = .workflow/active/WFS-{session}/.brainstorming/
 
     CHECK: brainstorm_dir/guidance-specification.md
     IF EXISTS:
@@ -88,13 +88,13 @@ Execute ui-designer analysis for existing topic framework
 
 ## Context Loading
 ASSIGNED_ROLE: ui-designer
-OUTPUT_LOCATION: .workflow/WFS-{session}/.brainstorming/ui-designer/
+OUTPUT_LOCATION: .workflow/active/WFS-{session}/.brainstorming/ui-designer/
 ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 ## Flow Control Steps
 1. **load_topic_framework**
    - Action: Load structured topic discussion framework
-   - Command: Read(.workflow/WFS-{session}/.brainstorming/guidance-specification.md)
+   - Command: Read(.workflow/active/WFS-{session}/.brainstorming/guidance-specification.md)
    - Output: topic_framework_content
 
 2. **load_role_template**
@@ -104,7 +104,7 @@ ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 3. **load_session_metadata**
    - Action: Load session metadata and existing context
-   - Command: Read(.workflow/WFS-{session}/workflow-session.json)
+   - Command: Read(.workflow/active/WFS-{session}/workflow-session.json)
    - Output: session_context
 
 ## Analysis Requirements
@@ -164,7 +164,7 @@ TodoWrite({
 
 ### Framework-Based Analysis
 ```
-.workflow/WFS-{session}/.brainstorming/ui-designer/
+.workflow/active/WFS-{session}/.brainstorming/ui-designer/
 └── analysis.md    # Structured analysis addressing guidance-specification.md discussion points
 ```
 
@@ -209,7 +209,7 @@ TodoWrite({
   "ui_designer": {
     "status": "completed",
     "framework_addressed": true,
-    "output_location": ".workflow/WFS-{session}/.brainstorming/ui-designer/analysis.md",
+    "output_location": ".workflow/active/WFS-{session}/.brainstorming/ui-designer/analysis.md",
     "framework_reference": "@../guidance-specification.md"
   }
 }

.claude/commands/workflow/brainstorm/ux-expert.md

@@ -1,6 +1,6 @@
 ---
 name: ux-expert
-description: Generate or update ux-expert/analysis.md addressing guidance-specification discussion points
+description: Generate or update ux-expert/analysis.md addressing guidance-specification discussion points for UX perspective
 argument-hint: "optional topic - uses existing framework if available"
 allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ---
@@ -48,10 +48,10 @@ allowed-tools: Task(conceptual-planning-agent), TodoWrite(*), Read(*), Write(*)
 ### Phase 1: Session & Framework Detection
 ```bash
 # Check active session and framework
-CHECK: .workflow/.active-* marker files
+CHECK: find .workflow/active/ -name "WFS-*" -type d
 IF active_session EXISTS:
     session_id = get_active_session()
-    brainstorm_dir = .workflow/WFS-{session}/.brainstorming/
+    brainstorm_dir = .workflow/active/WFS-{session}/.brainstorming/
 
     CHECK: brainstorm_dir/guidance-specification.md
     IF EXISTS:
@@ -88,13 +88,13 @@ Execute ux-expert analysis for existing topic framework
 
 ## Context Loading
 ASSIGNED_ROLE: ux-expert
-OUTPUT_LOCATION: .workflow/WFS-{session}/.brainstorming/ux-expert/
+OUTPUT_LOCATION: .workflow/active/WFS-{session}/.brainstorming/ux-expert/
 ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 ## Flow Control Steps
 1. **load_topic_framework**
    - Action: Load structured topic discussion framework
-   - Command: Read(.workflow/WFS-{session}/.brainstorming/guidance-specification.md)
+   - Command: Read(.workflow/active/WFS-{session}/.brainstorming/guidance-specification.md)
    - Output: topic_framework_content
 
 2. **load_role_template**
@@ -104,7 +104,7 @@ ANALYSIS_MODE: {framework_mode ? "framework_based" : "standalone"}
 
 3. **load_session_metadata**
    - Action: Load session metadata and existing context
-   - Command: Read(.workflow/WFS-{session}/workflow-session.json)
+   - Command: Read(.workflow/active/WFS-{session}/workflow-session.json)
    - Output: session_context
 
 ## Analysis Requirements
@@ -164,7 +164,7 @@ TodoWrite({
 
 ### Framework-Based Analysis
 ```
-.workflow/WFS-{session}/.brainstorming/ux-expert/
+.workflow/active/WFS-{session}/.brainstorming/ux-expert/
 └── analysis.md    # Structured analysis addressing guidance-specification.md discussion points
 ```
 
@@ -209,7 +209,7 @@ TodoWrite({
   "ux_expert": {
     "status": "completed",
     "framework_addressed": true,
-    "output_location": ".workflow/WFS-{session}/.brainstorming/ux-expert/analysis.md",
+    "output_location": ".workflow/active/WFS-{session}/.brainstorming/ux-expert/analysis.md",
     "framework_reference": "@../guidance-specification.md"
   }
 }

.claude/commands/workflow/clean.md

@@ -0,0 +1,516 @@
+---
+name: clean
+description: Intelligent code cleanup with mainline detection, stale artifact discovery, and safe execution
+argument-hint: "[--dry-run] [\"focus area\"]"
+allowed-tools: TodoWrite(*), Task(*), AskUserQuestion(*), Read(*), Glob(*), Bash(*), Write(*)
+---
+
+# Clean Command (/workflow:clean)
+
+## Overview
+
+Intelligent cleanup command that explores the codebase to identify the development mainline, discovers artifacts that have drifted from it, and safely removes stale sessions, abandoned documents, and dead code.
+
+**Core capabilities:**
+- Mainline detection: Identify active development branches and core modules
+- Drift analysis: Find sessions, documents, and code that deviate from mainline
+- Intelligent discovery: cli-explore-agent based artifact scanning
+- Safe execution: Confirmation-based cleanup with dry-run preview
+
+## Usage
+
+```bash
+/workflow:clean                          # Full intelligent cleanup (explore → analyze → confirm → execute)
+/workflow:clean --dry-run                # Explore and analyze only, no execution
+/workflow:clean "auth module"            # Focus cleanup on specific area
+```
+
+## Execution Process
+
+```
+Phase 1: Mainline Detection
+   ├─ Analyze git history for development trends
+   ├─ Identify core modules (high commit frequency)
+   ├─ Map active vs stale branches
+   └─ Build mainline profile
+
+Phase 2: Drift Discovery (cli-explore-agent)
+   ├─ Scan workflow sessions for orphaned artifacts
+   ├─ Identify documents drifted from mainline
+   ├─ Detect dead code and unused exports
+   └─ Generate cleanup manifest
+
+Phase 3: Confirmation
+   ├─ Display cleanup summary by category
+   ├─ Show impact analysis (files, size, risk)
+   └─ AskUserQuestion: Select categories to clean
+
+Phase 4: Execution (unless --dry-run)
+   ├─ Execute cleanup by category
+   ├─ Update manifests and indexes
+   └─ Report results
+```
+
+## Implementation
+
+### Phase 1: Mainline Detection
+
+**Session Setup**:
+```javascript
+const getUtc8ISOString = () => new Date(Date.now() + 8 * 60 * 60 * 1000).toISOString()
+
+const dateStr = getUtc8ISOString().substring(0, 10)
+const sessionId = `clean-${dateStr}`
+const sessionFolder = `.workflow/.clean/${sessionId}`
+
+Bash(`mkdir -p ${sessionFolder}`)
+```
+
+**Step 1.1: Git History Analysis**
+```bash
+# Get commit frequency by directory (last 30 days)
+bash(git log --since="30 days ago" --name-only --pretty=format: | grep -v "^$" | cut -d/ -f1-2 | sort | uniq -c | sort -rn | head -20)
+
+# Get recent active branches
+bash(git for-each-ref --sort=-committerdate refs/heads/ --format='%(refname:short) %(committerdate:relative)' | head -10)
+
+# Get files with most recent changes
+bash(git log --since="7 days ago" --name-only --pretty=format: | grep -v "^$" | sort | uniq -c | sort -rn | head -30)
+```
+
+**Step 1.2: Build Mainline Profile**
+```javascript
+const mainlineProfile = {
+  coreModules: [],        // High-frequency directories
+  activeFiles: [],        // Recently modified files
+  activeBranches: [],     // Branches with recent commits
+  staleThreshold: {
+    sessions: 7,          // Days
+    branches: 30,
+    documents: 14
+  },
+  timestamp: getUtc8ISOString()
+}
+
+// Parse git log output to identify core modules
+// Modules with >5 commits in last 30 days = core
+// Modules with 0 commits in last 30 days = potentially stale
+
+Write(`${sessionFolder}/mainline-profile.json`, JSON.stringify(mainlineProfile, null, 2))
+```
+
+---
+
+### Phase 2: Drift Discovery
+
+**Launch cli-explore-agent for intelligent artifact scanning**:
+
+```javascript
+Task(
+  subagent_type="cli-explore-agent",
+  run_in_background=false,
+  description="Discover stale artifacts",
+  prompt=`
+## Task Objective
+Discover artifacts that have drifted from the development mainline. Identify stale sessions, abandoned documents, and dead code for cleanup.
+
+## Context
+- **Session Folder**: ${sessionFolder}
+- **Mainline Profile**: ${sessionFolder}/mainline-profile.json
+- **Focus Area**: ${focusArea || "全项目"}
+
+## Discovery Categories
+
+### Category 1: Stale Workflow Sessions
+Scan and analyze workflow session directories:
+
+**Locations to scan**:
+- .workflow/active/WFS-* (active sessions)
+- .workflow/archives/WFS-* (archived sessions)
+- .workflow/.lite-plan/* (lite-plan sessions)
+- .workflow/.debug/DBG-* (debug sessions)
+
+**Staleness criteria**:
+- Active sessions: No modification >7 days + no related git commits
+- Archives: >30 days old + no feature references in project.json
+- Lite-plan: >7 days old + plan.json not executed
+- Debug: >3 days old + issue not in recent commits
+
+**Analysis steps**:
+1. List all session directories with modification times
+2. Cross-reference with git log (are session topics in recent commits?)
+3. Check manifest.json for orphan entries
+4. Identify sessions with .archiving marker (interrupted)
+
+### Category 2: Drifted Documents
+Scan documentation that no longer aligns with code:
+
+**Locations to scan**:
+- .claude/rules/tech/* (generated tech rules)
+- .workflow/.scratchpad/* (temporary notes)
+- **/CLAUDE.md (module documentation)
+- **/README.md (outdated descriptions)
+
+**Drift criteria**:
+- Tech rules: Referenced files no longer exist
+- Scratchpad: Any file (always temporary)
+- Module docs: Describe functions/classes that were removed
+- READMEs: Reference deleted directories
+
+**Analysis steps**:
+1. Parse document content for file/function references
+2. Verify referenced entities still exist in codebase
+3. Flag documents with >30% broken references
+
+### Category 3: Dead Code
+Identify code that is no longer used:
+
+**Scan patterns**:
+- Unused exports (exported but never imported)
+- Orphan files (not imported anywhere)
+- Commented-out code blocks (>10 lines)
+- TODO/FIXME comments >90 days old
+
+**Analysis steps**:
+1. Build import graph using rg/grep
+2. Identify exports with no importers
+3. Find files not in import graph
+4. Scan for large comment blocks
+
+## Output Format
+
+Write to: ${sessionFolder}/cleanup-manifest.json
+
+\`\`\`json
+{
+  "generated_at": "ISO timestamp",
+  "mainline_summary": {
+    "core_modules": ["src/core", "src/api"],
+    "active_branches": ["main", "feature/auth"],
+    "health_score": 0.85
+  },
+  "discoveries": {
+    "stale_sessions": [
+      {
+        "path": ".workflow/active/WFS-old-feature",
+        "type": "active",
+        "age_days": 15,
+        "reason": "No related commits in 15 days",
+        "size_kb": 1024,
+        "risk": "low"
+      }
+    ],
+    "drifted_documents": [
+      {
+        "path": ".claude/rules/tech/deprecated-lib",
+        "type": "tech_rules",
+        "broken_references": 5,
+        "total_references": 6,
+        "drift_percentage": 83,
+        "reason": "Referenced library removed",
+        "risk": "low"
+      }
+    ],
+    "dead_code": [
+      {
+        "path": "src/utils/legacy.ts",
+        "type": "orphan_file",
+        "reason": "Not imported by any file",
+        "last_modified": "2025-10-01",
+        "risk": "medium"
+      }
+    ]
+  },
+  "summary": {
+    "total_items": 12,
+    "total_size_mb": 45.2,
+    "by_category": {
+      "stale_sessions": 5,
+      "drifted_documents": 4,
+      "dead_code": 3
+    },
+    "by_risk": {
+      "low": 8,
+      "medium": 3,
+      "high": 1
+    }
+  }
+}
+\`\`\`
+
+## Execution Commands
+
+\`\`\`bash
+# Session directories
+find .workflow -type d -name "WFS-*" -o -name "DBG-*" 2>/dev/null
+
+# Check modification times (Linux/Mac)
+stat -c "%Y %n" .workflow/active/WFS-* 2>/dev/null
+
+# Check modification times (Windows PowerShell via bash)
+powershell -Command "Get-ChildItem '.workflow/active/WFS-*' | ForEach-Object { Write-Output \"$($_.LastWriteTime) $($_.FullName)\" }"
+
+# Find orphan exports (TypeScript)
+rg "export (const|function|class|interface|type)" --type ts -l
+
+# Find imports
+rg "import.*from" --type ts
+
+# Find large comment blocks
+rg "^\\s*/\\*" -A 10 --type ts
+
+# Find old TODOs
+rg "TODO|FIXME" --type ts -n
+\`\`\`
+
+## Success Criteria
+- [ ] All session directories scanned with age calculation
+- [ ] Documents cross-referenced with existing code
+- [ ] Dead code detection via import graph analysis
+- [ ] cleanup-manifest.json written with complete data
+- [ ] Each item has risk level and cleanup reason
+`
+)
+```
+
+---
+
+### Phase 3: Confirmation
+
+**Step 3.1: Display Summary**
+```javascript
+const manifest = JSON.parse(Read(`${sessionFolder}/cleanup-manifest.json`))
+
+console.log(`
+## Cleanup Discovery Report
+
+**Mainline Health**: ${Math.round(manifest.mainline_summary.health_score * 100)}%
+**Core Modules**: ${manifest.mainline_summary.core_modules.join(', ')}
+
+### Summary
+| Category | Count | Size | Risk |
+|----------|-------|------|------|
+| Stale Sessions | ${manifest.summary.by_category.stale_sessions} | - | ${getRiskSummary('sessions')} |
+| Drifted Documents | ${manifest.summary.by_category.drifted_documents} | - | ${getRiskSummary('documents')} |
+| Dead Code | ${manifest.summary.by_category.dead_code} | - | ${getRiskSummary('code')} |
+
+**Total**: ${manifest.summary.total_items} items, ~${manifest.summary.total_size_mb} MB
+
+### Stale Sessions
+${manifest.discoveries.stale_sessions.map(s =>
+  `- ${s.path} (${s.age_days}d, ${s.risk}): ${s.reason}`
+).join('\n')}
+
+### Drifted Documents
+${manifest.discoveries.drifted_documents.map(d =>
+  `- ${d.path} (${d.drift_percentage}% broken, ${d.risk}): ${d.reason}`
+).join('\n')}
+
+### Dead Code
+${manifest.discoveries.dead_code.map(c =>
+  `- ${c.path} (${c.type}, ${c.risk}): ${c.reason}`
+).join('\n')}
+`)
+```
+
+**Step 3.2: Dry-Run Exit**
+```javascript
+if (flags.includes('--dry-run')) {
+  console.log(`
+---
+**Dry-run mode**: No changes made.
+Manifest saved to: ${sessionFolder}/cleanup-manifest.json
+
+To execute cleanup: /workflow:clean
+`)
+  return
+}
+```
+
+**Step 3.3: User Confirmation**
+```javascript
+AskUserQuestion({
+  questions: [
+    {
+      question: "Which categories to clean?",
+      header: "Categories",
+      multiSelect: true,
+      options: [
+        {
+          label: "Sessions",
+          description: `${manifest.summary.by_category.stale_sessions} stale workflow sessions`
+        },
+        {
+          label: "Documents",
+          description: `${manifest.summary.by_category.drifted_documents} drifted documents`
+        },
+        {
+          label: "Dead Code",
+          description: `${manifest.summary.by_category.dead_code} unused code files`
+        }
+      ]
+    },
+    {
+      question: "Risk level to include?",
+      header: "Risk",
+      multiSelect: false,
+      options: [
+        { label: "Low only", description: "Safest - only obviously stale items" },
+        { label: "Low + Medium", description: "Recommended - includes likely unused items" },
+        { label: "All", description: "Aggressive - includes high-risk items" }
+      ]
+    }
+  ]
+})
+```
+
+---
+
+### Phase 4: Execution
+
+**Step 4.1: Filter Items by Selection**
+```javascript
+const selectedCategories = userSelection.categories  // ['Sessions', 'Documents', ...]
+const riskLevel = userSelection.risk                 // 'Low only', 'Low + Medium', 'All'
+
+const riskFilter = {
+  'Low only': ['low'],
+  'Low + Medium': ['low', 'medium'],
+  'All': ['low', 'medium', 'high']
+}[riskLevel]
+
+const itemsToClean = []
+
+if (selectedCategories.includes('Sessions')) {
+  itemsToClean.push(...manifest.discoveries.stale_sessions.filter(s => riskFilter.includes(s.risk)))
+}
+if (selectedCategories.includes('Documents')) {
+  itemsToClean.push(...manifest.discoveries.drifted_documents.filter(d => riskFilter.includes(d.risk)))
+}
+if (selectedCategories.includes('Dead Code')) {
+  itemsToClean.push(...manifest.discoveries.dead_code.filter(c => riskFilter.includes(c.risk)))
+}
+
+TodoWrite({
+  todos: itemsToClean.map(item => ({
+    content: `Clean: ${item.path}`,
+    status: "pending",
+    activeForm: `Cleaning ${item.path}`
+  }))
+})
+```
+
+**Step 4.2: Execute Cleanup**
+```javascript
+const results = { deleted: [], failed: [], skipped: [] }
+
+for (const item of itemsToClean) {
+  TodoWrite({ todos: [...] })  // Mark current as in_progress
+
+  try {
+    if (item.type === 'orphan_file' || item.type === 'dead_export') {
+      // Dead code: Delete file or remove export
+      Bash({ command: `rm -rf "${item.path}"` })
+    } else {
+      // Sessions and documents: Delete directory/file
+      Bash({ command: `rm -rf "${item.path}"` })
+    }
+
+    results.deleted.push(item.path)
+    TodoWrite({ todos: [...] })  // Mark as completed
+  } catch (error) {
+    results.failed.push({ path: item.path, error: error.message })
+  }
+}
+```
+
+**Step 4.3: Update Manifests**
+```javascript
+// Update archives manifest if sessions were deleted
+if (selectedCategories.includes('Sessions')) {
+  const archiveManifestPath = '.workflow/archives/manifest.json'
+  if (fileExists(archiveManifestPath)) {
+    const archiveManifest = JSON.parse(Read(archiveManifestPath))
+    const deletedSessionIds = results.deleted
+      .filter(p => p.includes('WFS-'))
+      .map(p => p.split('/').pop())
+
+    const updatedManifest = archiveManifest.filter(entry =>
+      !deletedSessionIds.includes(entry.session_id)
+    )
+
+    Write(archiveManifestPath, JSON.stringify(updatedManifest, null, 2))
+  }
+}
+
+// Update project.json if features referenced deleted sessions
+const projectPath = '.workflow/project.json'
+if (fileExists(projectPath)) {
+  const project = JSON.parse(Read(projectPath))
+  const deletedPaths = new Set(results.deleted)
+
+  project.features = project.features.filter(f =>
+    !deletedPaths.has(f.traceability?.archive_path)
+  )
+
+  project.statistics.total_features = project.features.length
+  project.statistics.last_updated = getUtc8ISOString()
+
+  Write(projectPath, JSON.stringify(project, null, 2))
+}
+```
+
+**Step 4.4: Report Results**
+```javascript
+console.log(`
+## Cleanup Complete
+
+**Deleted**: ${results.deleted.length} items
+**Failed**: ${results.failed.length} items
+**Skipped**: ${results.skipped.length} items
+
+### Deleted Items
+${results.deleted.map(p => `- ${p}`).join('\n')}
+
+${results.failed.length > 0 ? `
+### Failed Items
+${results.failed.map(f => `- ${f.path}: ${f.error}`).join('\n')}
+` : ''}
+
+Cleanup manifest archived to: ${sessionFolder}/cleanup-manifest.json
+`)
+```
+
+---
+
+## Session Folder Structure
+
+```
+.workflow/.clean/{YYYY-MM-DD}/
+├── mainline-profile.json     # Git history analysis
+└── cleanup-manifest.json     # Discovery results
+```
+
+## Risk Level Definitions
+
+| Risk | Description | Examples |
+|------|-------------|----------|
+| **Low** | Safe to delete, no dependencies | Empty sessions, scratchpad files, 100% broken docs |
+| **Medium** | Likely unused, verify before delete | Orphan files, old archives, partially broken docs |
+| **High** | May have hidden dependencies | Files with some imports, recent modifications |
+
+## Error Handling
+
+| Situation | Action |
+|-----------|--------|
+| No git repository | Skip mainline detection, use file timestamps only |
+| Session in use (.archiving) | Skip with warning |
+| Permission denied | Report error, continue with others |
+| Manifest parse error | Regenerate from filesystem scan |
+| Empty discovery | Report "codebase is clean" |
+
+## Related Commands
+
+- `/workflow:session:complete` - Properly archive active sessions
+- `/memory:compact` - Save session memory before cleanup
+- `/workflow:status` - View current workflow state

.claude/commands/workflow/debug.md

@@ -0,0 +1,321 @@
+---
+name: debug
+description: Interactive hypothesis-driven debugging with NDJSON logging, iterative until resolved
+argument-hint: "\"bug description or error message\""
+allowed-tools: TodoWrite(*), Task(*), AskUserQuestion(*), Read(*), Grep(*), Glob(*), Bash(*), Edit(*), Write(*)
+---
+
+# Workflow Debug Command (/workflow:debug)
+
+## Overview
+
+Evidence-based interactive debugging command. Systematically identifies root causes through hypothesis-driven logging and iterative verification.
+
+**Core workflow**: Explore → Add Logging → Reproduce → Analyze Log → Fix → Verify
+
+## Usage
+
+```bash
+/workflow:debug <BUG_DESCRIPTION>
+
+# Arguments
+<bug-description>          Bug description, error message, or stack trace (required)
+```
+
+## Execution Process
+
+```
+Session Detection:
+   ├─ Check if debug session exists for this bug
+   ├─ EXISTS + debug.log has content → Analyze mode
+   └─ NOT_FOUND or empty log → Explore mode
+
+Explore Mode:
+   ├─ Locate error source in codebase
+   ├─ Generate testable hypotheses (dynamic count)
+   ├─ Add NDJSON logging instrumentation
+   └─ Output: Hypothesis list + await user reproduction
+
+Analyze Mode:
+   ├─ Parse debug.log, validate each hypothesis
+   └─ Decision:
+       ├─ Confirmed → Fix root cause
+       ├─ Inconclusive → Add more logging, iterate
+       └─ All rejected → Generate new hypotheses
+
+Fix & Cleanup:
+   ├─ Apply fix based on confirmed hypothesis
+   ├─ User verifies
+   ├─ Remove debug instrumentation
+   └─ If not fixed → Return to Analyze mode
+```
+
+## Implementation
+
+### Session Setup & Mode Detection
+
+```javascript
+const getUtc8ISOString = () => new Date(Date.now() + 8 * 60 * 60 * 1000).toISOString()
+
+const bugSlug = bug_description.toLowerCase().replace(/[^a-z0-9]+/g, '-').substring(0, 30)
+const dateStr = getUtc8ISOString().substring(0, 10)
+
+const sessionId = `DBG-${bugSlug}-${dateStr}`
+const sessionFolder = `.workflow/.debug/${sessionId}`
+const debugLogPath = `${sessionFolder}/debug.log`
+
+// Auto-detect mode
+const sessionExists = fs.existsSync(sessionFolder)
+const logHasContent = sessionExists && fs.existsSync(debugLogPath) && fs.statSync(debugLogPath).size > 0
+
+const mode = logHasContent ? 'analyze' : 'explore'
+
+if (!sessionExists) {
+  bash(`mkdir -p ${sessionFolder}`)
+}
+```
+
+---
+
+### Explore Mode
+
+**Step 1.1: Locate Error Source**
+
+```javascript
+// Extract keywords from bug description
+const keywords = extractErrorKeywords(bug_description)
+// e.g., ['Stack Length', '未找到', 'registered 0']
+
+// Search codebase for error locations
+for (const keyword of keywords) {
+  Grep({ pattern: keyword, path: ".", output_mode: "content", "-C": 3 })
+}
+
+// Identify affected files and functions
+const affectedLocations = [...]  // from search results
+```
+
+**Step 1.2: Generate Hypotheses (Dynamic)**
+
+```javascript
+// Hypothesis categories based on error pattern
+const HYPOTHESIS_PATTERNS = {
+  "not found|missing|undefined|未找到": "data_mismatch",
+  "0|empty|zero|registered 0": "logic_error",
+  "timeout|connection|sync": "integration_issue",
+  "type|format|parse": "type_mismatch"
+}
+
+// Generate hypotheses based on actual issue (NOT fixed count)
+function generateHypotheses(bugDescription, affectedLocations) {
+  const hypotheses = []
+
+  // Analyze bug and create targeted hypotheses
+  // Each hypothesis has:
+  // - id: H1, H2, ... (dynamic count)
+  // - description: What might be wrong
+  // - testable_condition: What to log
+  // - logging_point: Where to add instrumentation
+
+  return hypotheses  // Could be 1, 3, 5, or more
+}
+
+const hypotheses = generateHypotheses(bug_description, affectedLocations)
+```
+
+**Step 1.3: Add NDJSON Instrumentation**
+
+For each hypothesis, add logging at the relevant location:
+
+**Python template**:
+```python
+# region debug [H{n}]
+try:
+    import json, time
+    _dbg = {
+        "sid": "{sessionId}",
+        "hid": "H{n}",
+        "loc": "{file}:{line}",
+        "msg": "{testable_condition}",
+        "data": {
+            # Capture relevant values here
+        },
+        "ts": int(time.time() * 1000)
+    }
+    with open(r"{debugLogPath}", "a", encoding="utf-8") as _f:
+        _f.write(json.dumps(_dbg, ensure_ascii=False) + "\n")
+except: pass
+# endregion
+```
+
+**JavaScript/TypeScript template**:
+```javascript
+// region debug [H{n}]
+try {
+  require('fs').appendFileSync("{debugLogPath}", JSON.stringify({
+    sid: "{sessionId}",
+    hid: "H{n}",
+    loc: "{file}:{line}",
+    msg: "{testable_condition}",
+    data: { /* Capture relevant values */ },
+    ts: Date.now()
+  }) + "\n");
+} catch(_) {}
+// endregion
+```
+
+**Output to user**:
+```
+## Hypotheses Generated
+
+Based on error "{bug_description}", generated {n} hypotheses:
+
+{hypotheses.map(h => `
+### ${h.id}: ${h.description}
+- Logging at: ${h.logging_point}
+- Testing: ${h.testable_condition}
+`).join('')}
+
+**Debug log**: ${debugLogPath}
+
+**Next**: Run reproduction steps, then come back for analysis.
+```
+
+---
+
+### Analyze Mode
+
+```javascript
+// Parse NDJSON log
+const entries = Read(debugLogPath).split('\n')
+  .filter(l => l.trim())
+  .map(l => JSON.parse(l))
+
+// Group by hypothesis
+const byHypothesis = groupBy(entries, 'hid')
+
+// Validate each hypothesis
+for (const [hid, logs] of Object.entries(byHypothesis)) {
+  const hypothesis = hypotheses.find(h => h.id === hid)
+  const latestLog = logs[logs.length - 1]
+
+  // Check if evidence confirms or rejects hypothesis
+  const verdict = evaluateEvidence(hypothesis, latestLog.data)
+  // Returns: 'confirmed' | 'rejected' | 'inconclusive'
+}
+```
+
+**Output**:
+```
+## Evidence Analysis
+
+Analyzed ${entries.length} log entries.
+
+${results.map(r => `
+### ${r.id}: ${r.description}
+- **Status**: ${r.verdict}
+- **Evidence**: ${JSON.stringify(r.evidence)}
+- **Reason**: ${r.reason}
+`).join('')}
+
+${confirmedHypothesis ? `
+## Root Cause Identified
+
+**${confirmedHypothesis.id}**: ${confirmedHypothesis.description}
+
+Ready to fix.
+` : `
+## Need More Evidence
+
+Add more logging or refine hypotheses.
+`}
+```
+
+---
+
+### Fix & Cleanup
+
+```javascript
+// Apply fix based on confirmed hypothesis
+// ... Edit affected files
+
+// After user verifies fix works:
+
+// Remove debug instrumentation (search for region markers)
+const instrumentedFiles = Grep({
+  pattern: "# region debug|// region debug",
+  output_mode: "files_with_matches"
+})
+
+for (const file of instrumentedFiles) {
+  // Remove content between region markers
+  removeDebugRegions(file)
+}
+
+console.log(`
+## Debug Complete
+
+- Root cause: ${confirmedHypothesis.description}
+- Fix applied to: ${modifiedFiles.join(', ')}
+- Debug instrumentation removed
+`)
+```
+
+---
+
+## Debug Log Format (NDJSON)
+
+Each line is a JSON object:
+
+```json
+{"sid":"DBG-xxx-2025-12-18","hid":"H1","loc":"file.py:func:42","msg":"Check dict keys","data":{"keys":["a","b"],"target":"c","found":false},"ts":1734567890123}
+```
+
+| Field | Description |
+|-------|-------------|
+| `sid` | Session ID |
+| `hid` | Hypothesis ID (H1, H2, ...) |
+| `loc` | Code location |
+| `msg` | What's being tested |
+| `data` | Captured values |
+| `ts` | Timestamp (ms) |
+
+## Session Folder
+
+```
+.workflow/.debug/DBG-{slug}-{date}/
+├── debug.log          # NDJSON log (main artifact)
+└── resolution.md      # Summary after fix (optional)
+```
+
+## Iteration Flow
+
+```
+First Call (/workflow:debug "error"):
+   ├─ No session exists → Explore mode
+   ├─ Extract error keywords, search codebase
+   ├─ Generate hypotheses, add logging
+   └─ Await user reproduction
+
+After Reproduction (/workflow:debug "error"):
+   ├─ Session exists + debug.log has content → Analyze mode
+   ├─ Parse log, evaluate hypotheses
+   └─ Decision:
+       ├─ Confirmed → Fix → User verify
+       │   ├─ Fixed → Cleanup → Done
+       │   └─ Not fixed → Add logging → Iterate
+       ├─ Inconclusive → Add logging → Iterate
+       └─ All rejected → New hypotheses → Iterate
+
+Output:
+   └─ .workflow/.debug/DBG-{slug}-{date}/debug.log
+```
+
+## Error Handling
+
+| Situation | Action |
+|-----------|--------|
+| Empty debug.log | Verify reproduction triggered the code path |
+| All hypotheses rejected | Generate new hypotheses with broader scope |
+| Fix doesn't work | Iterate with more granular logging |
+| >5 iterations | Escalate to `/workflow:lite-fix` with evidence |

.claude/commands/workflow/execute.md

@@ -1,13 +1,13 @@
 ---
 name: execute
-description: Coordinate agents for existing workflow tasks with automatic discovery
+description: Coordinate agent execution for workflow tasks with automatic session discovery, parallel task processing, and status tracking
 argument-hint: "[--resume-session=\"session-id\"]"
 ---
 
 # Workflow Execute Command
 
 ## Overview
-Orchestrates autonomous workflow execution through systematic task discovery, agent coordination, and progress tracking. **Executes entire workflow without user interruption**, providing complete context to agents and ensuring proper flow control execution with comprehensive TodoWrite tracking.
+Orchestrates autonomous workflow execution through systematic task discovery, agent coordination, and progress tracking. **Executes entire workflow without user interruption** (except initial session selection if multiple active sessions exist), providing complete context to agents and ensuring proper flow control execution with comprehensive TodoWrite tracking.
 
 **Resume Mode**: When called with `--resume-session` flag, skips discovery phase and directly enters TodoWrite generation and agent execution for the specified session.
 
@@ -15,130 +15,240 @@ Orchestrates autonomous workflow execution through systematic task discovery, ag
 
 **Lazy Loading**: Task JSONs read **on-demand** during execution, not upfront. TODO_LIST.md + IMPL_PLAN.md provide metadata for planning.
 
-| Metric | Before | After | Improvement |
-|--------|--------|-------|-------------|
-| **Initial Load** | All task JSONs (~2,300 lines) | TODO_LIST.md only (~650 lines) | **72% reduction** |
-| **Startup Time** | Seconds | Milliseconds | **~90% faster** |
-| **Memory** | All tasks | 1-2 tasks | **90% less** |
-| **Scalability** | 10-20 tasks | 100+ tasks | **5-10x** |
+**Loading Strategy**:
+- **TODO_LIST.md**: Read in Phase 3 (task metadata, status, dependencies for TodoWrite generation)
+- **IMPL_PLAN.md**: Check existence in Phase 2 (normal mode), parse execution strategy in Phase 4A
+- **Task JSONs**: Lazy loading - read only when task is about to execute (Phase 4B)
 
 ## Core Rules
 **Complete entire workflow autonomously without user interruption, using TodoWrite for comprehensive progress tracking.**
-**Execute all discovered pending tasks sequentially until workflow completion or blocking dependency.**
+**Execute all discovered pending tasks until workflow completion or blocking dependency.**
 **Auto-complete session when all tasks finished: Call `/workflow:session:complete` upon workflow completion.**
+**ONE AGENT = ONE TASK JSON: Each agent instance executes exactly one task JSON file - never batch multiple tasks into single agent execution.**
 
 ## Core Responsibilities
 - **Session Discovery**: Identify and select active workflow sessions
-- **Task Dependency Resolution**: Analyze task relationships and execution order
+- **Execution Strategy Parsing**: Extract execution model from IMPL_PLAN.md
 - **TodoWrite Progress Tracking**: Maintain real-time execution status throughout entire workflow
 - **Agent Orchestration**: Coordinate specialized agents with complete context
-- **Flow Control Execution**: Execute pre-analysis steps and context accumulation
 - **Status Synchronization**: Update task JSON files and workflow state
 - **Autonomous Completion**: Continue execution until all tasks complete or reach blocking state
 - **Session Auto-Complete**: Call `/workflow:session:complete` when all workflow tasks finished
 
 ## Execution Philosophy
-- **Discovery-first**: Auto-discover existing plans and tasks
-- **Status-aware**: Execute only ready tasks with resolved dependencies
-- **Context-rich**: Provide complete task JSON and accumulated context to agents
-- **Progress tracking**: **Continuous TodoWrite updates throughout entire workflow execution**
-- **Flow control**: Sequential step execution with variable passing
-- **Autonomous completion**: **Execute all tasks without user interruption until workflow complete**
+- **Progress tracking**: Continuous TodoWrite updates throughout entire workflow execution
+- **Autonomous completion**: Execute all tasks without user interruption until workflow complete
 
-## Flow Control Execution
-**[FLOW_CONTROL]** marker indicates task JSON contains `flow_control.pre_analysis` steps for context preparation.
+## Execution Process
 
-### Orchestrator Responsibility
-- Pass complete task JSON to agent (including `flow_control` block)
-- Provide session paths for artifact access
-- Monitor agent completion
+```
+Normal Mode:
+Phase 1: Discovery
+   ├─ Count active sessions
+   └─ Decision:
+      ├─ count=0 → ERROR: No active sessions
+      ├─ count=1 → Auto-select session → Phase 2
+      └─ count>1 → AskUserQuestion (max 4 options) → Phase 2
 
-### Agent Responsibility
-- Parse `flow_control.pre_analysis` array from JSON
-- Execute steps sequentially with variable substitution
-- Accumulate context from artifacts and dependencies
-- Follow error handling per `step.on_error`
-- Complete implementation using accumulated context
+Phase 2: Planning Document Validation
+   ├─ Check IMPL_PLAN.md exists
+   ├─ Check TODO_LIST.md exists
+   └─ Validate .task/ contains IMPL-*.json files
 
-**Orchestrator does NOT execute flow control steps - Agent interprets and executes them from JSON.**
+Phase 3: TodoWrite Generation
+   ├─ Update session status to "active" (Step 0)
+   ├─ Parse TODO_LIST.md for task statuses
+   ├─ Generate TodoWrite for entire workflow
+   └─ Prepare session context paths
+
+Phase 4: Execution Strategy & Task Execution
+   ├─ Step 4A: Parse execution strategy from IMPL_PLAN.md
+   └─ Step 4B: Execute tasks with lazy loading
+      └─ Loop:
+         ├─ Get next in_progress task from TodoWrite
+         ├─ Lazy load task JSON
+         ├─ Launch agent with task context
+         ├─ Mark task completed (update IMPL-*.json status)
+         │  # Quick fix: Update task status for ccw dashboard
+         │  # TS=$(date -Iseconds) && jq --arg ts "$TS" '.status="completed" | .status_history=(.status_history // [])+[{"from":"in_progress","to":"completed","changed_at":$ts}]' IMPL-X.json > tmp.json && mv tmp.json IMPL-X.json
+         └─ Advance to next task
+
+Phase 5: Completion
+   ├─ Update task statuses in JSON files
+   ├─ Generate summaries
+   └─ Auto-call /workflow:session:complete
+
+Resume Mode (--resume-session):
+   ├─ Skip Phase 1 & Phase 2
+   └─ Entry Point: Phase 3 (TodoWrite Generation)
+      ├─ Update session status to "active" (if not already)
+      └─ Continue: Phase 4 → Phase 5
+```
 
 ## Execution Lifecycle
 
-### Resume Mode Detection
-**Special Flag Processing**: When `--resume-session="session-id"` is provided:
-1. **Skip Discovery Phase**: Use provided session ID directly
-2. **Load Specified Session**: Read session state from `.workflow/{session-id}/`
-3. **Direct TodoWrite Generation**: Skip to Phase 3 (Planning) immediately
-4. **Accelerated Execution**: Enter agent coordination without validation delays
+### Phase 1: Discovery
+**Applies to**: Normal mode only (skipped in resume mode)
 
-### Phase 1: Discovery (Normal Mode Only)
-1. **Check Active Sessions**: Find `.workflow/.active-*` markers
-2. **Select Session**: If multiple found, prompt user selection
-3. **Load Session Metadata**: Read `workflow-session.json` ONLY (minimal context)
-4. **DO NOT read task JSONs yet** - defer until execution phase
+**Purpose**: Find and select active workflow session with user confirmation when multiple sessions exist
 
-**Note**: In resume mode, this phase is completely skipped.
+**Process**:
 
-### Phase 2: Planning Document Analysis (Normal Mode Only)
-**Optimized to avoid reading all task JSONs upfront**
+#### Step 1.1: Count Active Sessions
+```bash
+bash(find .workflow/active/ -name "WFS-*" -type d 2>/dev/null | wc -l)
+```
 
-1. **Read IMPL_PLAN.md**: Understand overall strategy, task breakdown summary, dependencies
-2. **Read TODO_LIST.md**: Get current task statuses and execution progress
-3. **Extract Task Metadata**: Parse task IDs, titles, and dependency relationships from TODO_LIST.md
-4. **Build Execution Queue**: Determine ready tasks based on TODO_LIST.md status and dependencies
+#### Step 1.2: Handle Session Selection
 
-**Key Optimization**: Use IMPL_PLAN.md and TODO_LIST.md as primary sources instead of reading all task JSONs
+**Case A: No Sessions** (count = 0)
+```
+ERROR: No active workflow sessions found
+Run /workflow:plan "task description" to create a session
+```
 
-**Note**: In resume mode, this phase is also skipped as session analysis was already completed by `/workflow:status`.
+**Case B: Single Session** (count = 1)
+```bash
+bash(find .workflow/active/ -name "WFS-*" -type d 2>/dev/null | head -1 | xargs basename)
+```
+Auto-select and continue to Phase 2.
 
-### Phase 3: TodoWrite Generation (Resume Mode Entry Point)
-**This is where resume mode directly enters after skipping Phases 1 & 2**
+**Case C: Multiple Sessions** (count > 1)
 
+List sessions with metadata and prompt user selection:
+```bash
+bash(for dir in .workflow/active/WFS-*/; do
+  session=$(basename "$dir")
+  project=$(jq -r '.project // "Unknown"' "$dir/workflow-session.json" 2>/dev/null)
+  total=$(grep -c "^- \[" "$dir/TODO_LIST.md" 2>/dev/null || echo "0")
+  completed=$(grep -c "^- \[x\]" "$dir/TODO_LIST.md" 2>/dev/null || echo "0")
+  [ "$total" -gt 0 ] && progress=$((completed * 100 / total)) || progress=0
+  echo "${session} | ${project} | ${completed}/${total} tasks (${progress}%)"
+done)
+```
+
+Use AskUserQuestion to present formatted options (max 4 options shown):
+```javascript
+// If more than 4 sessions, show most recent 4 with "Other" option for manual input
+const sessions = getActiveSessions()  // sorted by last modified
+const displaySessions = sessions.slice(0, 4)
+
+AskUserQuestion({
+  questions: [{
+    question: "Multiple active sessions detected. Select one:",
+    header: "Session",
+    multiSelect: false,
+    options: displaySessions.map(s => ({
+      label: s.id,
+      description: `${s.project} | ${s.progress}`
+    }))
+    // Note: User can select "Other" to manually enter session ID
+  }]
+})
+```
+
+**Input Validation**:
+- If user selects from options: Use selected session ID
+- If user selects "Other" and provides input: Validate session exists
+- If validation fails: Show error and re-prompt or suggest available sessions
+
+Parse user input (supports: number "1", full ID "WFS-auth-system", or partial "auth"), validate selection, and continue to Phase 2.
+
+#### Step 1.3: Load Session Metadata
+```bash
+bash(cat .workflow/active/${sessionId}/workflow-session.json)
+```
+
+**Output**: Store session metadata in memory
+**DO NOT read task JSONs yet** - defer until execution phase (lazy loading)
+
+**Resume Mode**: This entire phase is skipped when `--resume-session="session-id"` flag is provided.
+
+### Phase 2: Planning Document Validation
+**Applies to**: Normal mode only (skipped in resume mode)
+
+**Purpose**: Validate planning artifacts exist before execution
+
+**Process**:
+1. **Check IMPL_PLAN.md**: Verify file exists (defer detailed parsing to Phase 4A)
+2. **Check TODO_LIST.md**: Verify file exists (defer reading to Phase 3)
+3. **Validate Task Directory**: Ensure `.task/` contains at least one IMPL-*.json file
+
+**Key Optimization**: Only existence checks here. Actual file reading happens in later phases.
+
+**Resume Mode**: This phase is skipped when `--resume-session` flag is provided. Resume mode entry point is Phase 3.
+
+### Phase 3: TodoWrite Generation
+**Applies to**: Both normal and resume modes (resume mode entry point)
+
+**Step 0: Update Session Status to Active**
+Before generating TodoWrite, update session status from "planning" to "active":
+```bash
+# Update session status (idempotent - safe to run if already active)
+jq '.status = "active" | .execution_started_at = (.execution_started_at // now | todate)' \
+  .workflow/active/${sessionId}/workflow-session.json > tmp.json && \
+  mv tmp.json .workflow/active/${sessionId}/workflow-session.json
+```
+This ensures the dashboard shows the session as "ACTIVE" during execution.
+
+**Process**:
 1. **Create TodoWrite List**: Generate task list from TODO_LIST.md (not from task JSONs)
    - Parse TODO_LIST.md to extract all tasks with current statuses
    - Identify first pending task with met dependencies
    - Generate comprehensive TodoWrite covering entire workflow
-2. **Mark Initial Status**: Set first ready task as `in_progress` in TodoWrite
-3. **Prepare Session Context**: Inject workflow paths for agent use (using provided session-id)
-4. **Validate Prerequisites**: Ensure IMPL_PLAN.md and TODO_LIST.md exist and are valid
+2. **Prepare Session Context**: Inject workflow paths for agent use (using provided session-id)
+3. **Validate Prerequisites**: Ensure IMPL_PLAN.md and TODO_LIST.md exist and are valid
 
 **Resume Mode Behavior**:
-- Load existing TODO_LIST.md directly from `.workflow/{session-id}/`
+- Load existing TODO_LIST.md directly from `.workflow/active/{session-id}/`
 - Extract current progress from TODO_LIST.md
 - Generate TodoWrite from TODO_LIST.md state
 - Proceed immediately to agent execution (Phase 4)
 
-### Phase 4: Execution (Lazy Task Loading)
-**Key Optimization**: Read task JSON **only when needed** for execution
+### Phase 4: Execution Strategy Selection & Task Execution
+**Applies to**: Both normal and resume modes
 
-1. **Identify Next Task**: From TodoWrite, get the next `in_progress` task ID
-2. **Load Task JSON on Demand**: Read `.task/{task-id}.json` for current task ONLY
-3. **Validate Task Structure**: Ensure all 5 required fields exist (id, title, status, meta, context, flow_control)
-4. **Pass Task with Flow Control**: Include complete task JSON with `pre_analysis` steps for agent execution
-5. **Launch Agent**: Invoke specialized agent with complete context including flow control steps
-6. **Monitor Progress**: Track agent execution and handle errors without user interruption
-7. **Collect Results**: Gather implementation results and outputs
-8. **Update TODO_LIST.md**: Mark current task as completed in TODO_LIST.md
-9. **Continue Workflow**: Identify next pending task from TODO_LIST.md and repeat from step 1
+**Step 4A: Parse Execution Strategy from IMPL_PLAN.md**
+
+Read IMPL_PLAN.md Section 4 to extract:
+- **Execution Model**: Sequential | Parallel | Phased | TDD Cycles
+- **Parallelization Opportunities**: Which tasks can run in parallel
+- **Serialization Requirements**: Which tasks must run sequentially
+- **Critical Path**: Priority execution order
+
+If IMPL_PLAN.md lacks execution strategy, use intelligent fallback (analyze task structure).
+
+**Step 4B: Execute Tasks with Lazy Loading**
+
+**Key Optimization**: Read task JSON **only when needed** for execution
 
 **Execution Loop Pattern**:
 ```
 while (TODO_LIST.md has pending tasks) {
   next_task_id = getTodoWriteInProgressTask()
-  task_json = Read(.workflow/{session}/.task/{next_task_id}.json)  // Lazy load
+  task_json = Read(.workflow/active/{session}/.task/{next_task_id}.json)  // Lazy load
   executeTaskWithAgent(task_json)
   updateTodoListMarkCompleted(next_task_id)
   advanceTodoWriteToNextTask()
 }
 ```
 
-**Benefits**:
-- Reduces initial context loading by ~90%
-- Only reads task JSON when actually executing
-- Scales better for workflows with many tasks
-- Faster startup time for workflow execution
+**Execution Process per Task**:
+1. **Identify Next Task**: From TodoWrite, get the next `in_progress` task ID
+2. **Load Task JSON on Demand**: Read `.task/{task-id}.json` for current task ONLY
+3. **Validate Task Structure**: Ensure all 5 required fields exist (id, title, status, meta, context, flow_control)
+4. **Launch Agent**: Invoke specialized agent with complete context including flow control steps
+5. **Monitor Progress**: Track agent execution and handle errors without user interruption
+6. **Collect Results**: Gather implementation results and outputs
+7. **Continue Workflow**: Identify next pending task from TODO_LIST.md and repeat
+
+**Note**: TODO_LIST.md updates are handled by agents (e.g., code-developer.md), not by the orchestrator.
+
 
 ### Phase 5: Completion
+**Applies to**: Both normal and resume modes
+
+**Process**:
 1. **Update Task Status**: Mark completed tasks in JSON files
 2. **Generate Summary**: Create task summary in `.summaries/`
 3. **Update TodoWrite**: Mark current task complete, advance to next
@@ -146,34 +256,53 @@ while (TODO_LIST.md has pending tasks) {
 5. **Check Workflow Complete**: Verify all tasks are completed
 6. **Auto-Complete Session**: Call `/workflow:session:complete` when all tasks finished
 
-## Task Discovery & Queue Building
+## Execution Strategy (IMPL_PLAN-Driven)
 
-### Session Discovery Process (Normal Mode - Optimized)
-```
-├── Check for .active-* markers in .workflow/
-├── If multiple active sessions found → Prompt user to select
-├── Locate selected session's workflow folder
-├── Load session metadata: workflow-session.json (minimal context)
-├── Read IMPL_PLAN.md (strategy overview and task summary)
-├── Read TODO_LIST.md (current task statuses and dependencies)
-├── Parse TODO_LIST.md to extract task metadata (NO JSON loading)
-├── Build execution queue from TODO_LIST.md
-└── Generate TodoWrite from TODO_LIST.md state
-```
+### Strategy Priority
 
-**Key Change**: Task JSONs are NOT loaded during discovery - they are loaded lazily during execution
+**IMPL_PLAN-Driven Execution (Recommended)**:
+1. **Read IMPL_PLAN.md execution strategy** (Section 4: Implementation Strategy)
+2. **Follow explicit guidance**:
+   - Execution Model (Sequential/Parallel/Phased/TDD)
+   - Parallelization Opportunities (which tasks can run in parallel)
+   - Serialization Requirements (which tasks must run sequentially)
+   - Critical Path (priority execution order)
+3. **Use TODO_LIST.md for status tracking** only
+4. **IMPL_PLAN decides "HOW"**, execute.md implements it
 
-### Resume Mode Process (--resume-session flag - Optimized)
-```
-├── Use provided session-id directly (skip discovery)
-├── Validate .workflow/{session-id}/ directory exists
-├── Read TODO_LIST.md for current progress
-├── Parse TODO_LIST.md to extract task IDs and statuses
-├── Generate TodoWrite from TODO_LIST.md (prioritize in-progress/pending tasks)
-└── Enter Phase 4 (Execution) with lazy task JSON loading
-```
+**Intelligent Fallback (When IMPL_PLAN lacks execution details)**:
+1. **Analyze task structure**:
+   - Check `meta.execution_group` in task JSONs
+   - Analyze `depends_on` relationships
+   - Understand task complexity and risk
+2. **Apply smart defaults**:
+   - No dependencies + same execution_group → Parallel
+   - Has dependencies → Sequential (wait for deps)
+   - Critical/high-risk tasks → Sequential
+3. **Conservative approach**: When uncertain, prefer sequential execution
 
-**Key Change**: Completely skip IMPL_PLAN.md and task JSON loading - use TODO_LIST.md only
+### Execution Models
+
+#### 1. Sequential Execution
+**When**: IMPL_PLAN specifies "Sequential" OR no clear parallelization guidance
+**Pattern**: Execute tasks one by one in TODO_LIST order
+**TodoWrite**: ONE task marked as `in_progress` at a time
+
+#### 2. Parallel Execution
+**When**: IMPL_PLAN specifies "Parallel" with clear parallelization opportunities
+**Pattern**: Execute independent task groups concurrently by launching multiple agent instances
+**TodoWrite**: MULTIPLE tasks (in same batch) marked as `in_progress` simultaneously
+**Agent Instantiation**: Launch one agent instance per task (respects ONE AGENT = ONE TASK JSON rule)
+
+#### 3. Phased Execution
+**When**: IMPL_PLAN specifies "Phased" with phase breakdown
+**Pattern**: Execute tasks in phases, respect phase boundaries
+**TodoWrite**: Within each phase, follow Sequential or Parallel rules
+
+#### 4. Intelligent Fallback
+**When**: IMPL_PLAN lacks execution strategy details
+**Pattern**: Analyze task structure and apply smart defaults
+**TodoWrite**: Follow Sequential or Parallel rules based on analysis
 
 ### Task Status Logic
 ```
@@ -182,155 +311,36 @@ completed → skip
 blocked → skip until dependencies clear
 ```
 
-## Batch Execution with Dependency Graph
-
-### Parallel Execution Algorithm
-**Core principle**: Execute independent tasks concurrently in batches based on dependency graph.
-
-#### Algorithm Steps (Optimized with Lazy Loading)
-```javascript
-function executeBatchWorkflow(sessionId) {
-  // 1. Build dependency graph from TODO_LIST.md (NOT task JSONs)
-  const graph = buildDependencyGraphFromTodoList(`.workflow/${sessionId}/TODO_LIST.md`);
-
-  // 2. Process batches until graph is empty
-  while (!graph.isEmpty()) {
-    // 3. Identify current batch (tasks with in-degree = 0)
-    const batch = graph.getNodesWithInDegreeZero();
-
-    // 4. Load task JSONs ONLY for current batch (lazy loading)
-    const batchTaskJsons = batch.map(taskId =>
-      Read(`.workflow/${sessionId}/.task/${taskId}.json`)
-    );
-
-    // 5. Check for parallel execution opportunities
-    const parallelGroups = groupByExecutionGroup(batchTaskJsons);
-
-    // 6. Execute batch concurrently
-    await Promise.all(
-      parallelGroups.map(group => executeBatch(group))
-    );
-
-    // 7. Update graph: remove completed tasks and their edges
-    graph.removeNodes(batch);
-
-    // 8. Update TODO_LIST.md and TodoWrite to reflect completed batch
-    updateTodoListAfterBatch(batch);
-    updateTodoWriteAfterBatch(batch);
-  }
-
-  // 9. All tasks complete - auto-complete session
-  SlashCommand("/workflow:session:complete");
-}
-
-function buildDependencyGraphFromTodoList(todoListPath) {
-  const todoContent = Read(todoListPath);
-  const tasks = parseTodoListTasks(todoContent);
-  const graph = new DirectedGraph();
-
-  tasks.forEach(task => {
-    graph.addNode(task.id, { id: task.id, title: task.title, status: task.status });
-    task.dependencies?.forEach(depId => graph.addEdge(depId, task.id));
-  });
-
-  return graph;
-}
-
-function parseTodoListTasks(todoContent) {
-  // Parse: - [ ] **IMPL-001**: Task title → [📋](./.task/IMPL-001.json)
-  const taskPattern = /- \[([ x])\] \*\*([A-Z]+-\d+(?:\.\d+)?)\*\*: (.+?) →/g;
-  const tasks = [];
-  let match;
-
-  while ((match = taskPattern.exec(todoContent)) !== null) {
-    tasks.push({
-      status: match[1] === 'x' ? 'completed' : 'pending',
-      id: match[2],
-      title: match[3]
-    });
-  }
-
-  return tasks;
-}
-
-function groupByExecutionGroup(tasks) {
-  const groups = {};
-
-  tasks.forEach(task => {
-    const groupId = task.meta.execution_group || task.id;
-    if (!groups[groupId]) groups[groupId] = [];
-    groups[groupId].push(task);
-  });
-
-  return Object.values(groups);
-}
-
-async function executeBatch(tasks) {
-  // Execute all tasks in batch concurrently
-  return Promise.all(
-    tasks.map(task => executeTask(task))
-  );
-}
-```
-
-#### Execution Group Rules
-1. **Same `execution_group` ID** → Execute in parallel (independent, different contexts)
-2. **No `execution_group` (null)** → Execute sequentially (has dependencies)
-3. **Different `execution_group` IDs** → Execute in parallel (independent batches)
-4. **Same `context_signature`** → Should have been merged (warning if not)
-
-#### Parallel Execution Example
-```
-Batch 1 (no dependencies):
-  - IMPL-1.1 (execution_group: "parallel-auth-api") → Agent 1
-  - IMPL-1.2 (execution_group: "parallel-ui-comp") → Agent 2
-  - IMPL-1.3 (execution_group: "parallel-db-schema") → Agent 3
-
-Wait for Batch 1 completion...
-
-Batch 2 (depends on Batch 1):
-  - IMPL-2.1 (execution_group: null, depends_on: [IMPL-1.1, IMPL-1.2]) → Agent 1
-
-Wait for Batch 2 completion...
-
-Batch 3 (independent of Batch 2):
-  - IMPL-3.1 (execution_group: "parallel-tests-1") → Agent 1
-  - IMPL-3.2 (execution_group: "parallel-tests-2") → Agent 2
-```
-
 ## TodoWrite Coordination
-**Comprehensive workflow tracking** with immediate status updates throughout entire execution without user interruption:
 
-#### TodoWrite Workflow Rules
-1. **Initial Creation**: Generate TodoWrite from discovered pending tasks for entire workflow
-   - **Normal Mode**: Create from discovery results
-   - **Resume Mode**: Create from existing session state and current progress
-2. **Parallel Task Support**:
-   - **Single-task execution**: Mark ONLY ONE task as `in_progress` at a time
-   - **Batch execution**: Mark ALL tasks in current batch as `in_progress` simultaneously
-   - **Execution group indicator**: Show `[execution_group: group-id]` for parallel tasks
-3. **Immediate Updates**: Update status after each task/batch completion without user interruption
-4. **Status Synchronization**: Sync with JSON task files after updates
-5. **Continuous Tracking**: Maintain TodoWrite throughout entire workflow execution until completion
+### TodoWrite Rules (Unified)
 
-#### Resume Mode TodoWrite Generation
-**Special behavior when `--resume-session` flag is present**:
-- Load existing session progress from `.workflow/{session-id}/TODO_LIST.md`
-- Identify currently in-progress or next pending task
-- Generate TodoWrite starting from interruption point
-- Preserve completed task history in TodoWrite display
-- Focus on remaining pending tasks for execution
+**Rule 1: Initial Creation**
+- **Normal Mode**: Generate TodoWrite from discovered pending tasks for entire workflow
+- **Resume Mode**: Generate from existing session state and current progress
 
-#### TodoWrite Tool Usage
-**Use Claude Code's built-in TodoWrite tool** to track workflow progress in real-time:
+**Rule 2: In-Progress Task Count (Execution-Model-Dependent)**
+- **Sequential execution**: Mark ONLY ONE task as `in_progress` at a time
+- **Parallel batch execution**: Mark ALL tasks in current batch as `in_progress` simultaneously
+- **Execution group indicator**: Show `[execution_group: group-id]` for parallel tasks
 
+**Rule 3: Status Updates**
+- **Immediate Updates**: Update status after each task/batch completion without user interruption
+- **Status Synchronization**: Sync with JSON task files after updates
+- **Continuous Tracking**: Maintain TodoWrite throughout entire workflow execution until completion
+
+**Rule 4: Workflow Completion Check**
+- When all tasks marked `completed`, auto-call `/workflow:session:complete`
+
+### TodoWrite Tool Usage
+
+**Example 1: Sequential Execution**
 ```javascript
-// Example 1: Sequential execution (traditional)
 TodoWrite({
   todos: [
     {
       content: "Execute IMPL-1.1: Design auth schema [code-developer] [FLOW_CONTROL]",
-      status: "in_progress",  // Single task in progress
+      status: "in_progress",  // ONE task in progress
       activeForm: "Executing IMPL-1.1: Design auth schema"
     },
     {
@@ -340,8 +350,10 @@ TodoWrite({
     }
   ]
 });
+```
 
-// Example 2: Batch execution (parallel tasks with execution_group)
+**Example 2: Parallel Batch Execution**
+```javascript
 TodoWrite({
   todos: [
     {
@@ -366,255 +378,50 @@ TodoWrite({
     }
   ]
 });
-
-// Example 3: After batch completion
-TodoWrite({
-  todos: [
-    {
-      content: "Execute IMPL-1.1: Build Auth API [code-developer] [execution_group: parallel-auth-api]",
-      status: "completed",  // Batch completed
-      activeForm: "Executing IMPL-1.1: Build Auth API"
-    },
-    {
-      content: "Execute IMPL-1.2: Build User UI [code-developer] [execution_group: parallel-ui-comp]",
-      status: "completed",  // Batch completed
-      activeForm: "Executing IMPL-1.2: Build User UI"
-    },
-    {
-      content: "Execute IMPL-1.3: Setup Database [code-developer] [execution_group: parallel-db-schema]",
-      status: "completed",  // Batch completed
-      activeForm: "Executing IMPL-1.3: Setup Database"
-    },
-    {
-      content: "Execute IMPL-2.1: Integration Tests [test-fix-agent]",
-      status: "in_progress",  // Next batch started
-      activeForm: "Executing IMPL-2.1: Integration Tests"
-    }
-  ]
-});
 ```
 
-**TodoWrite Integration Rules**:
-- **Continuous Workflow Tracking**: Use TodoWrite tool throughout entire workflow execution
-- **Real-time Updates**: Immediate progress tracking without user interruption
-- **Single Active Task**: Only ONE task marked as `in_progress` at any time
-- **Immediate Completion**: Mark tasks `completed` immediately after finishing
-- **Status Sync**: Sync TodoWrite status with JSON task files after each update
-- **Full Execution**: Continue TodoWrite tracking until all workflow tasks complete
-- **Workflow Completion Check**: When all tasks marked `completed`, auto-call `/workflow:session:complete`
+## Agent Execution Pattern
 
-#### TODO_LIST.md Update Timing
-**Single source of truth for task status** - enables lazy loading by providing task metadata without reading JSONs
+### Flow Control Execution
+**[FLOW_CONTROL]** marker indicates task JSON contains `flow_control.pre_analysis` steps for context preparation.
 
-- **Before Agent Launch**: Mark task as `in_progress`
-- **After Task Complete**: Mark as `completed`, advance to next
-- **On Error**: Keep as `in_progress`, add error note
-- **Workflow Complete**: Call `/workflow:session:complete`
+**Note**: Orchestrator does NOT execute flow control steps - Agent interprets and executes them autonomously.
 
-### 3. Agent Context Management
-**Comprehensive context preparation** for autonomous agent execution:
+### Agent Prompt Template
+**Dynamic Generation**: Before agent invocation, read task JSON and extract key requirements.
 
-#### Context Sources (Priority Order)
-1. **Complete Task JSON**: Full task definition including all fields and artifacts
-2. **Artifacts Context**: Brainstorming outputs and role analysess from task.context.artifacts
-3. **Flow Control Context**: Accumulated outputs from pre_analysis steps (including artifact loading)
-4. **Dependency Summaries**: Previous task completion summaries
-5. **Session Context**: Workflow paths and session metadata
-6. **Inherited Context**: Parent task context and shared variables
-
-#### Context Assembly Process
-```
-1. Load Task JSON → Base context (including artifacts array)
-2. Load Artifacts → Synthesis specifications and brainstorming outputs
-3. Execute Flow Control → Accumulated context (with artifact loading steps)
-4. Load Dependencies → Dependency context
-5. Prepare Session Paths → Session context
-6. Combine All → Complete agent context with artifact integration
-```
-
-#### Agent Context Package Structure
-```json
-{
-  "task": { /* Complete task JSON with artifacts array */ },
-  "artifacts": {
-    "synthesis_specification": { "path": "{{from context-package.json → brainstorm_artifacts.synthesis_output.path}}", "priority": "highest" },
-    "guidance_specification": { "path": "{{from context-package.json → brainstorm_artifacts.guidance_specification.path}}", "priority": "medium" },
-    "role_analyses": [ /* From context-package.json → brainstorm_artifacts.role_analyses[] */ ],
-    "conflict_resolution": { "path": "{{from context-package.json → brainstorm_artifacts.conflict_resolution.path}}", "conditional": true }
-  },
-  "flow_context": {
-    "step_outputs": {
-      "synthesis_specification": "...",
-      "individual_artifacts": "...",
-      "pattern_analysis": "...",
-      "dependency_context": "..."
-    }
-  },
-  "session": {
-    "workflow_dir": ".workflow/WFS-session/",
-    "context_package_path": ".workflow/WFS-session/.process/context-package.json",
-    "todo_list_path": ".workflow/WFS-session/TODO_LIST.md",
-    "summaries_dir": ".workflow/WFS-session/.summaries/",
-    "task_json_path": ".workflow/WFS-session/.task/IMPL-1.1.json"
-  },
-  "dependencies": [ /* Task summaries from depends_on */ ],
-  "inherited": { /* Parent task context */ }
-}
-```
-
-#### Context Validation Rules
-- **Task JSON Complete**: All 5 fields present and valid, including artifacts array in context
-- **Artifacts Available**: All artifacts loaded from context-package.json
-- **Flow Control Ready**: All pre_analysis steps completed including artifact loading steps
-- **Dependencies Loaded**: All depends_on summaries available
-- **Session Paths Valid**: All workflow paths exist and accessible (verified via context-package.json)
-- **Agent Assignment**: Valid agent type specified in meta.agent
-
-### 4. Agent Execution Pattern
-**Structured agent invocation** with complete context and clear instructions:
-
-#### Agent Prompt Template
 ```bash
 Task(subagent_type="{meta.agent}",
-     prompt="**EXECUTE TASK FROM JSON**
+     run_in_background=false,
+     prompt="Execute task: {task.title}
 
-     ## Task JSON Location
-     {session.task_json_path}
+     {[FLOW_CONTROL]}
 
-     ## Instructions
-     1. **Load Complete Task JSON**: Read and validate all fields (id, title, status, meta, context, flow_control)
-     2. **Execute Flow Control**: If `flow_control.pre_analysis` exists, execute steps sequentially:
-        - Load artifacts (role analysis documents, role analyses) using commands in each step
-        - Accumulate context from step outputs using variable substitution [variable_name]
-        - Handle errors per step.on_error (skip_optional | fail | retry_once)
-     3. **Implement Solution**: Follow `flow_control.implementation_approach` using accumulated context
-     4. **Complete Task**:
-        - Update task status: `jq '.status = \"completed\"' {session.task_json_path} > temp.json && mv temp.json {session.task_json_path}`
-        - Update TODO_LIST.md: Mark task as [x] completed in {session.todo_list_path}
-        - Generate summary: {session.summaries_dir}/{task.id}-summary.md
-        - Check workflow completion and call `/workflow:session:complete` if all tasks done
+     **Task Objectives** (from task JSON):
+     {task.context.objective}
 
-     ## Context Sources (All from JSON)
-     - Requirements: `context.requirements`
-     - Focus Paths: `context.focus_paths`
-     - Acceptance: `context.acceptance`
-     - Artifacts: `context.artifacts` (synthesis specs, brainstorming outputs)
-     - Dependencies: `context.depends_on`
-     - Target Files: `flow_control.target_files`
+     **Expected Deliverables** (from task JSON):
+     {task.context.deliverables}
 
-     ## Session Paths
+     **Quality Standards** (from task JSON):
+     {task.context.acceptance_criteria}
+
+     **MANDATORY FIRST STEPS**:
+     1. Read complete task JSON: {session.task_json_path}
+     2. Load context package: {session.context_package_path}
+
+
+     **Session Paths**:
      - Workflow Dir: {session.workflow_dir}
      - TODO List: {session.todo_list_path}
-     - Summaries: {session.summaries_dir}
-     - Flow Context: {flow_context.step_outputs}
+     - Summaries Dir: {session.summaries_dir}
+     - Context Package: {session.context_package_path}
 
-     **Complete JSON structure is authoritative - load and follow it exactly.**"),
-     description="Execute task: {task.id}")
+     **Success Criteria**: Complete all objectives, meet all quality standards, deliver all outputs as specified above.",
+     description="Executing: {task.title}")
 ```
 
-#### Agent JSON Loading Specification
-**MANDATORY AGENT PROTOCOL**: All agents must follow this exact loading sequence:
-
-1. **JSON Loading**: First action must be `cat {session.task_json_path}`
-2. **Field Validation**: Verify all 5 required fields exist: `id`, `title`, `status`, `meta`, `context`, `flow_control`
-3. **Structure Parsing**: Parse nested fields correctly:
-   - `meta.type` and `meta.agent` (NOT flat `task_type`)
-   - `context.requirements`, `context.focus_paths`, `context.acceptance`
-   - `context.depends_on`, `context.inherited`
-   - `flow_control.pre_analysis` array, `flow_control.target_files`
-4. **Flow Control Execution**: If `flow_control.pre_analysis` exists, execute steps sequentially
-5. **Status Management**: Update JSON status upon completion
-
-**JSON Field Reference**:
-```json
-{
-  "id": "IMPL-1.2",
-  "title": "Task title",
-  "status": "pending|active|completed|blocked",
-  "meta": {
-    "type": "feature|bugfix|refactor|test-gen|test-fix|docs",
-    "agent": "@code-developer|@test-fix-agent|@universal-executor"
-  },
-  "context": {
-    "requirements": ["req1", "req2"],
-    "focus_paths": ["src/path1", "src/path2"],
-    "acceptance": ["criteria1", "criteria2"],
-    "depends_on": ["IMPL-1.1"],
-    "inherited": { "from": "parent", "context": ["info"] },
-    "artifacts": [
-      {
-        "type": "synthesis_specification",
-        "source": "context-package.json → brainstorm_artifacts.synthesis_output",
-        "path": "{{loaded dynamically from context-package.json}}",
-        "priority": "highest",
-        "contains": "complete_integrated_specification"
-      },
-      {
-        "type": "individual_role_analysis",
-        "source": "context-package.json → brainstorm_artifacts.role_analyses[]",
-        "path": "{{loaded dynamically from context-package.json}}",
-        "note": "Supports analysis*.md pattern (analysis.md, analysis-01.md, analysis-api.md, etc.)",
-        "priority": "low",
-        "contains": "role_specific_analysis_fallback"
-      }
-    ]
-  },
-  "flow_control": {
-    "pre_analysis": [
-      {
-        "step": "load_synthesis_specification",
-        "action": "Load synthesis specification from context-package.json",
-        "commands": [
-          "Read(.workflow/WFS-[session]/.process/context-package.json)",
-          "Extract(brainstorm_artifacts.synthesis_output.path)",
-          "Read(extracted path)"
-        ],
-        "output_to": "synthesis_specification",
-        "on_error": "skip_optional"
-      },
-      {
-        "step": "step_name",
-        "command": "bash_command",
-        "output_to": "variable",
-        "on_error": "skip_optional|fail|retry_once"
-      }
-    ],
-    "implementation_approach": [
-      {
-        "step": 1,
-        "title": "Implement task following role analyses",
-        "description": "Implement '[title]' following role analyses. PRIORITY: Use role analysis documents as primary requirement source. When implementation needs technical details (e.g., API schemas, caching configs, design tokens), refer to artifacts[] for detailed specifications from original role analyses.",
-        "modification_points": [
-          "Apply consolidated requirements from role analysis documents",
-          "Follow technical guidelines from synthesis",
-          "Consult artifacts for implementation details when needed",
-          "Integrate with existing patterns"
-        ],
-        "logic_flow": [
-          "Load role analyses",
-          "Parse architecture and requirements",
-          "Implement following specification",
-          "Consult artifacts for technical details when needed",
-          "Validate against acceptance criteria"
-        ],
-        "depends_on": [],
-        "output": "implementation"
-      }
-    ],
-    "target_files": ["file:function:lines", "path/to/NewFile.ts"]
-  }
-}
-```
-
-#### Execution Flow
-1. **Load Task JSON**: Agent reads and validates complete JSON structure
-2. **Execute Flow Control**: Agent runs pre_analysis steps if present
-3. **Prepare Implementation**: Agent uses implementation_approach from JSON
-4. **Launch Implementation**: Agent follows focus_paths and target_files
-5. **Update Status**: Agent marks JSON status as completed
-6. **Generate Summary**: Agent creates completion summary
-
-#### Agent Assignment Rules
+### Agent Assignment Rules
 ```
 meta.agent specified → Use specified agent
 meta.agent missing → Infer from meta.type:
@@ -625,18 +432,12 @@ meta.agent missing → Infer from meta.type:
   - "docs" → @doc-generator
 ```
 
-#### Error Handling During Execution
-- **Agent Failure**: Retry once with adjusted context
-- **Flow Control Error**: Skip optional steps, fail on critical
-- **Context Missing**: Reload from JSON files and retry
-- **Timeout**: Mark as blocked, continue with next task
-
 ## Workflow File Structure Reference
 ```
-.workflow/WFS-[topic-slug]/
+.workflow/active/WFS-[topic-slug]/
 ├── workflow-session.json     # Session state and metadata
 ├── IMPL_PLAN.md             # Planning document and requirements
-├── TODO_LIST.md             # Progress tracking (auto-updated)
+├── TODO_LIST.md             # Progress tracking (updated by agents)
 ├── .task/                   # Task definitions (JSON only)
 │   ├── IMPL-1.json          # Main task definitions
 │   └── IMPL-1.1.json        # Subtask definitions
@@ -644,78 +445,26 @@ meta.agent missing → Infer from meta.type:
 │   ├── IMPL-1-summary.md    # Task completion details
 │   └── IMPL-1.1-summary.md  # Subtask completion details
 └── .process/                # Planning artifacts
+    ├── context-package.json # Smart context package
     └── ANALYSIS_RESULTS.md  # Planning analysis results
 ```
 
 ## Error Handling & Recovery
 
-### Discovery Phase Errors
-| Error | Cause | Resolution | Command |
-|-------|-------|------------|---------|
-| No active session | No `.active-*` markers found | Create or resume session | `/workflow:plan "project"` |
-| Multiple sessions | Multiple `.active-*` markers | Select specific session | Manual choice prompt |
-| Corrupted session | Invalid JSON files | Recreate session structure | `/workflow:session:status --validate` |
-| Missing task files | Broken task references | Regenerate tasks | `/task:create` or repair |
+### Common Errors & Recovery
 
-### Execution Phase Errors
-| Error | Cause | Recovery Strategy | Max Attempts |
-|-------|-------|------------------|--------------|
+| Error Type | Cause | Recovery Strategy | Max Attempts |
+|-----------|-------|------------------|--------------|
+| **Discovery Errors** |
+| No active session | No sessions in `.workflow/active/` | Create or resume session: `/workflow:plan "project"` | N/A |
+| Multiple sessions | Multiple sessions in `.workflow/active/` | Prompt user selection | N/A |
+| Corrupted session | Invalid JSON files | Recreate session structure or validate files | N/A |
+| **Execution Errors** |
 | Agent failure | Agent crash/timeout | Retry with simplified context | 2 |
 | Flow control error | Command failure | Skip optional, fail critical | 1 per step |
 | Context loading error | Missing dependencies | Reload from JSON, use defaults | 3 |
 | JSON file corruption | File system issues | Restore from backup/recreate | 1 |
 
-### Recovery Procedures
-
-#### Session Recovery
-```bash
-# Check session integrity
-find .workflow -name ".active-*" | while read marker; do
-  session=$(basename "$marker" | sed 's/^\.active-//')
-  if [ ! -d ".workflow/$session" ]; then
-    echo "Removing orphaned marker: $marker"
-    rm "$marker"
-  fi
-done
-
-# Recreate corrupted session files
-if [ ! -f ".workflow/$session/workflow-session.json" ]; then
-  echo '{"session_id":"'$session'","status":"active"}' > ".workflow/$session/workflow-session.json"
-fi
-```
-
-#### Task Recovery
-```bash
-# Validate task JSON integrity
-for task_file in .workflow/$session/.task/*.json; do
-  if ! jq empty "$task_file" 2>/dev/null; then
-    echo "Corrupted task file: $task_file"
-    # Backup and regenerate or restore from backup
-  fi
-done
-
-# Fix missing dependencies
-missing_deps=$(jq -r '.context.depends_on[]?' .workflow/$session/.task/*.json | sort -u)
-for dep in $missing_deps; do
-  if [ ! -f ".workflow/$session/.task/$dep.json" ]; then
-    echo "Missing dependency: $dep - creating placeholder"
-  fi
-done
-```
-
-#### Context Recovery
-```bash
-# Reload context from available sources
-if [ -f ".workflow/$session/.process/ANALYSIS_RESULTS.md" ]; then
-  echo "Reloading planning context..."
-fi
-
-# Restore from documentation if available
-if [ -d ".workflow/docs/" ]; then
-  echo "Using documentation context as fallback..."
-fi
-```
-
 ### Error Prevention
 - **Pre-flight Checks**: Validate session integrity before execution
 - **Backup Strategy**: Create task snapshots before major operations
@@ -723,16 +472,3 @@ fi
 - **Dependency Validation**: Check all depends_on references exist
 - **Context Verification**: Ensure all required context is available
 
-## Usage Examples
-
-### Basic Usage
-```bash
-/workflow:execute          # Execute all pending tasks autonomously
-/workflow:session:status           # Check progress
-/task:execute IMPL-1.2     # Execute specific task
-```
-
-### Integration
-- **Planning**: `/workflow:plan` → `/workflow:execute` → `/workflow:review`
-- **Recovery**: `/workflow:status --validate` → `/workflow:execute`
-

.claude/commands/workflow/init.md

@@ -0,0 +1,211 @@
+---
+name: init
+description: Initialize project-level state with intelligent project analysis using cli-explore-agent
+argument-hint: "[--regenerate]"
+examples:
+  - /workflow:init
+  - /workflow:init --regenerate
+---
+
+# Workflow Init Command (/workflow:init)
+
+## Overview
+Initialize `.workflow/project-tech.json` and `.workflow/project-guidelines.json` with comprehensive project understanding by delegating analysis to **cli-explore-agent**.
+
+**Dual File System**:
+- `project-tech.json`: Auto-generated technical analysis (stack, architecture, components)
+- `project-guidelines.json`: User-maintained rules and constraints (created as scaffold)
+
+**Note**: This command may be called by other workflow commands. Upon completion, return immediately to continue the calling workflow without interrupting the task flow.
+
+## Usage
+```bash
+/workflow:init                 # Initialize (skip if exists)
+/workflow:init --regenerate    # Force regeneration
+```
+
+## Execution Process
+
+```
+Input Parsing:
+   └─ Parse --regenerate flag → regenerate = true | false
+
+Decision:
+   ├─ BOTH_EXIST + no --regenerate → Exit: "Already initialized"
+   ├─ EXISTS + --regenerate → Backup existing → Continue analysis
+   └─ NOT_FOUND → Continue analysis
+
+Analysis Flow:
+   ├─ Get project metadata (name, root)
+   ├─ Invoke cli-explore-agent
+   │   ├─ Structural scan (get_modules_by_depth.sh, find, wc)
+   │   ├─ Semantic analysis (Gemini CLI)
+   │   ├─ Synthesis and merge
+   │   └─ Write .workflow/project-tech.json
+   ├─ Create guidelines scaffold (if not exists)
+   │   └─ Write .workflow/project-guidelines.json (empty structure)
+   └─ Display summary
+
+Output:
+   ├─ .workflow/project-tech.json (+ .backup if regenerate)
+   └─ .workflow/project-guidelines.json (scaffold if new)
+```
+
+## Implementation
+
+### Step 1: Parse Input and Check Existing State
+
+**Parse --regenerate flag**:
+```javascript
+const regenerate = $ARGUMENTS.includes('--regenerate')
+```
+
+**Check existing state**:
+
+```bash
+bash(test -f .workflow/project-tech.json && echo "TECH_EXISTS" || echo "TECH_NOT_FOUND")
+bash(test -f .workflow/project-guidelines.json && echo "GUIDELINES_EXISTS" || echo "GUIDELINES_NOT_FOUND")
+```
+
+**If BOTH_EXIST and no --regenerate**: Exit early
+```
+Project already initialized:
+- Tech analysis: .workflow/project-tech.json
+- Guidelines: .workflow/project-guidelines.json
+
+Use /workflow:init --regenerate to rebuild tech analysis
+Use /workflow:session:solidify to add guidelines
+Use /workflow:status --project to view state
+```
+
+### Step 2: Get Project Metadata
+
+```bash
+bash(basename "$(git rev-parse --show-toplevel 2>/dev/null || pwd)")
+bash(git rev-parse --show-toplevel 2>/dev/null || pwd)
+bash(mkdir -p .workflow)
+```
+
+### Step 3: Invoke cli-explore-agent
+
+**For --regenerate**: Backup and preserve existing data
+```bash
+bash(cp .workflow/project-tech.json .workflow/project-tech.json.backup)
+```
+
+**Delegate analysis to agent**:
+
+```javascript
+Task(
+  subagent_type="cli-explore-agent",
+  run_in_background=false,
+  description="Deep project analysis",
+  prompt=`
+Analyze project for workflow initialization and generate .workflow/project-tech.json.
+
+## MANDATORY FIRST STEPS
+1. Execute: cat ~/.claude/workflows/cli-templates/schemas/project-tech-schema.json (get schema reference)
+2. Execute: ccw tool exec get_modules_by_depth '{}' (get project structure)
+
+## Task
+Generate complete project-tech.json with:
+- project_metadata: {name: ${projectName}, root_path: ${projectRoot}, initialized_at, updated_at}
+- technology_analysis: {description, languages, frameworks, build_tools, test_frameworks, architecture, key_components, dependencies}
+- development_status: ${regenerate ? 'preserve from backup' : '{completed_features: [], development_index: {feature: [], enhancement: [], bugfix: [], refactor: [], docs: []}, statistics: {total_features: 0, total_sessions: 0, last_updated}}'}
+- _metadata: {initialized_by: "cli-explore-agent", analysis_timestamp, analysis_mode}
+
+## Analysis Requirements
+
+**Technology Stack**:
+- Languages: File counts, mark primary
+- Frameworks: From package.json, requirements.txt, go.mod, etc.
+- Build tools: npm, cargo, maven, webpack, vite
+- Test frameworks: jest, pytest, go test, junit
+
+**Architecture**:
+- Style: MVC, microservices, layered (from structure & imports)
+- Layers: presentation, business-logic, data-access
+- Patterns: singleton, factory, repository
+- Key components: 5-10 modules {name, path, description, importance}
+
+## Execution
+1. Structural scan: get_modules_by_depth.sh, find, wc -l
+2. Semantic analysis: Gemini for patterns/architecture
+3. Synthesis: Merge findings
+4. ${regenerate ? 'Merge with preserved development_status from .workflow/project-tech.json.backup' : ''}
+5. Write JSON: Write('.workflow/project-tech.json', jsonContent)
+6. Report: Return brief completion summary
+
+Project root: ${projectRoot}
+`
+)
+```
+
+### Step 3.5: Create Guidelines Scaffold (if not exists)
+
+```javascript
+// Only create if not exists (never overwrite user guidelines)
+if (!file_exists('.workflow/project-guidelines.json')) {
+  const guidelinesScaffold = {
+    conventions: {
+      coding_style: [],
+      naming_patterns: [],
+      file_structure: [],
+      documentation: []
+    },
+    constraints: {
+      architecture: [],
+      tech_stack: [],
+      performance: [],
+      security: []
+    },
+    quality_rules: [],
+    learnings: [],
+    _metadata: {
+      created_at: new Date().toISOString(),
+      version: "1.0.0"
+    }
+  };
+
+  Write('.workflow/project-guidelines.json', JSON.stringify(guidelinesScaffold, null, 2));
+}
+```
+
+### Step 4: Display Summary
+
+```javascript
+const projectTech = JSON.parse(Read('.workflow/project-tech.json'));
+const guidelinesExists = file_exists('.workflow/project-guidelines.json');
+
+console.log(`
+✓ Project initialized successfully
+
+## Project Overview
+Name: ${projectTech.project_metadata.name}
+Description: ${projectTech.technology_analysis.description}
+
+### Technology Stack
+Languages: ${projectTech.technology_analysis.languages.map(l => l.name).join(', ')}
+Frameworks: ${projectTech.technology_analysis.frameworks.join(', ')}
+
+### Architecture
+Style: ${projectTech.technology_analysis.architecture.style}
+Components: ${projectTech.technology_analysis.key_components.length} core modules
+
+---
+Files created:
+- Tech analysis: .workflow/project-tech.json
+- Guidelines: .workflow/project-guidelines.json ${guidelinesExists ? '(scaffold)' : ''}
+${regenerate ? '- Backup: .workflow/project-tech.json.backup' : ''}
+
+Next steps:
+- Use /workflow:session:solidify to add project guidelines
+- Use /workflow:plan to start planning
+`);
+```
+
+## Error Handling
+
+**Agent Failure**: Fall back to basic initialization with placeholder overview
+**Missing Tools**: Agent uses Qwen fallback or bash-only
+**Empty Project**: Create minimal JSON with all gaps identified

.claude/commands/workflow/lite-execute.md

@@ -0,0 +1,792 @@
+---
+name: lite-execute
+description: Execute tasks based on in-memory plan, prompt description, or file content
+argument-hint: "[--in-memory] [\"task description\"|file-path]"
+allowed-tools: TodoWrite(*), Task(*), Bash(*)
+---
+
+# Workflow Lite-Execute Command (/workflow:lite-execute)
+
+## Overview
+
+Flexible task execution command supporting three input modes: in-memory plan (from lite-plan), direct prompt description, or file content. Handles execution orchestration, progress tracking, and optional code review.
+
+**Core capabilities:**
+- Multi-mode input (in-memory plan, prompt description, or file path)
+- Execution orchestration (Agent or Codex) with full context
+- Live progress tracking via TodoWrite at execution call level
+- Optional code review with selected tool (Gemini, Agent, or custom)
+- Context continuity across multiple executions
+- Intelligent format detection (Enhanced Task JSON vs plain text)
+
+## Usage
+
+### Command Syntax
+```bash
+/workflow:lite-execute [FLAGS] <INPUT>
+
+# Flags
+--in-memory                Use plan from memory (called by lite-plan)
+
+# Arguments
+<input>                    Task description string, or path to file (required)
+```
+
+## Input Modes
+
+### Mode 1: In-Memory Plan
+
+**Trigger**: Called by lite-plan after Phase 4 approval with `--in-memory` flag
+
+**Input Source**: `executionContext` global variable set by lite-plan
+
+**Content**: Complete execution context (see Data Structures section)
+
+**Behavior**:
+- Skip execution method selection (already set by lite-plan)
+- Directly proceed to execution with full context
+- All planning artifacts available (exploration, clarifications, plan)
+
+### Mode 2: Prompt Description
+
+**Trigger**: User calls with task description string
+
+**Input**: Simple task description (e.g., "Add unit tests for auth module")
+
+**Behavior**:
+- Store prompt as `originalUserInput`
+- Create simple execution plan from prompt
+- AskUserQuestion: Select execution method (Agent/Codex/Auto)
+- AskUserQuestion: Select code review tool (Skip/Gemini/Agent/Other)
+- Proceed to execution with `originalUserInput` included
+
+**User Interaction**:
+```javascript
+AskUserQuestion({
+  questions: [
+    {
+      question: "Select execution method:",
+      header: "Execution",
+      multiSelect: false,
+      options: [
+        { label: "Agent", description: "@code-developer agent" },
+        { label: "Codex", description: "codex CLI tool" },
+        { label: "Auto", description: "Auto-select based on complexity" }
+      ]
+    },
+    {
+      question: "Enable code review after execution?",
+      header: "Code Review",
+      multiSelect: false,
+      options: [
+        { label: "Skip", description: "No review" },
+        { label: "Gemini Review", description: "Gemini CLI tool" },
+        { label: "Agent Review", description: "Current agent review" }
+      ]
+    }
+  ]
+})
+```
+
+### Mode 3: File Content
+
+**Trigger**: User calls with file path
+
+**Input**: Path to file containing task description or plan.json
+
+**Step 1: Read and Detect Format**
+
+```javascript
+fileContent = Read(filePath)
+
+// Attempt JSON parsing
+try {
+  jsonData = JSON.parse(fileContent)
+
+  // Check if plan.json from lite-plan session
+  if (jsonData.summary && jsonData.approach && jsonData.tasks) {
+    planObject = jsonData
+    originalUserInput = jsonData.summary
+    isPlanJson = true
+  } else {
+    // Valid JSON but not plan.json - treat as plain text
+    originalUserInput = fileContent
+    isPlanJson = false
+  }
+} catch {
+  // Not valid JSON - treat as plain text prompt
+  originalUserInput = fileContent
+  isPlanJson = false
+}
+```
+
+**Step 2: Create Execution Plan**
+
+If `isPlanJson === true`:
+- Use `planObject` directly
+- User selects execution method and code review
+
+If `isPlanJson === false`:
+- Treat file content as prompt (same behavior as Mode 2)
+- Create simple execution plan from content
+
+**Step 3: User Interaction**
+
+- AskUserQuestion: Select execution method (Agent/Codex/Auto)
+- AskUserQuestion: Select code review tool
+- Proceed to execution with full context
+
+## Execution Process
+
+```
+Input Parsing:
+   └─ Decision (mode detection):
+      ├─ --in-memory flag → Mode 1: Load executionContext → Skip user selection
+      ├─ Ends with .md/.json/.txt → Mode 3: Read file → Detect format
+      │   ├─ Valid plan.json → Use planObject → User selects method + review
+      │   └─ Not plan.json → Treat as prompt → User selects method + review
+      └─ Other → Mode 2: Prompt description → User selects method + review
+
+Execution:
+   ├─ Step 1: Initialize result tracking (previousExecutionResults = [])
+   ├─ Step 2: Task grouping & batch creation
+   │   ├─ Extract explicit depends_on (no file/keyword inference)
+   │   ├─ Group: independent tasks → single parallel batch (maximize utilization)
+   │   ├─ Group: dependent tasks → sequential phases (respect dependencies)
+   │   └─ Create TodoWrite list for batches
+   ├─ Step 3: Launch execution
+   │   ├─ Phase 1: All independent tasks (⚡ single batch, concurrent)
+   │   └─ Phase 2+: Dependent tasks by dependency order
+   ├─ Step 4: Track progress (TodoWrite updates per batch)
+   └─ Step 5: Code review (if codeReviewTool ≠ "Skip")
+
+Output:
+   └─ Execution complete with results in previousExecutionResults[]
+```
+
+## Detailed Execution Steps
+
+### Step 1: Initialize Execution Tracking
+
+**Operations**:
+- Initialize result tracking for multi-execution scenarios
+- Set up `previousExecutionResults` array for context continuity
+
+```javascript
+// Initialize result tracking
+previousExecutionResults = []
+```
+
+### Step 2: Task Grouping & Batch Creation
+
+**Dependency Analysis & Grouping Algorithm**:
+```javascript
+// Use explicit depends_on from plan.json (no inference from file/keywords)
+function extractDependencies(tasks) {
+  const taskIdToIndex = {}
+  tasks.forEach((t, i) => { taskIdToIndex[t.id] = i })
+
+  return tasks.map((task, i) => {
+    // Only use explicit depends_on from plan.json
+    const deps = (task.depends_on || [])
+      .map(depId => taskIdToIndex[depId])
+      .filter(idx => idx !== undefined && idx < i)
+    return { ...task, taskIndex: i, dependencies: deps }
+  })
+}
+
+// Group into batches: maximize parallel execution
+function createExecutionCalls(tasks, executionMethod) {
+  const tasksWithDeps = extractDependencies(tasks)
+  const processed = new Set()
+  const calls = []
+
+  // Phase 1: All independent tasks → single parallel batch (maximize utilization)
+  const independentTasks = tasksWithDeps.filter(t => t.dependencies.length === 0)
+  if (independentTasks.length > 0) {
+    independentTasks.forEach(t => processed.add(t.taskIndex))
+    calls.push({
+      method: executionMethod,
+      executionType: "parallel",
+      groupId: "P1",
+      taskSummary: independentTasks.map(t => t.title).join(' | '),
+      tasks: independentTasks
+    })
+  }
+
+  // Phase 2: Dependent tasks → sequential batches (respect dependencies)
+  let sequentialIndex = 1
+  let remaining = tasksWithDeps.filter(t => !processed.has(t.taskIndex))
+
+  while (remaining.length > 0) {
+    // Find tasks whose dependencies are all satisfied
+    const ready = remaining.filter(t =>
+      t.dependencies.every(d => processed.has(d))
+    )
+
+    if (ready.length === 0) {
+      console.warn('Circular dependency detected, forcing remaining tasks')
+      ready.push(...remaining)
+    }
+
+    // Group ready tasks (can run in parallel within this phase)
+    ready.forEach(t => processed.add(t.taskIndex))
+    calls.push({
+      method: executionMethod,
+      executionType: ready.length > 1 ? "parallel" : "sequential",
+      groupId: ready.length > 1 ? `P${calls.length + 1}` : `S${sequentialIndex++}`,
+      taskSummary: ready.map(t => t.title).join(ready.length > 1 ? ' | ' : ' → '),
+      tasks: ready
+    })
+
+    remaining = remaining.filter(t => !processed.has(t.taskIndex))
+  }
+
+  return calls
+}
+
+executionCalls = createExecutionCalls(planObject.tasks, executionMethod).map(c => ({ ...c, id: `[${c.groupId}]` }))
+
+TodoWrite({
+  todos: executionCalls.map(c => ({
+    content: `${c.executionType === "parallel" ? "⚡" : "→"} ${c.id} (${c.tasks.length} tasks)`,
+    status: "pending",
+    activeForm: `Executing ${c.id}`
+  }))
+})
+```
+
+### Step 3: Launch Execution
+
+**Executor Resolution** (任务级 executor 优先于全局设置):
+```javascript
+// 获取任务的 executor（优先使用 executorAssignments，fallback 到全局 executionMethod）
+function getTaskExecutor(task) {
+  const assignments = executionContext?.executorAssignments || {}
+  if (assignments[task.id]) {
+    return assignments[task.id].executor  // 'gemini' | 'codex' | 'agent'
+  }
+  // Fallback: 全局 executionMethod 映射
+  const method = executionContext?.executionMethod || 'Auto'
+  if (method === 'Agent') return 'agent'
+  if (method === 'Codex') return 'codex'
+  // Auto: 根据复杂度
+  return planObject.complexity === 'Low' ? 'agent' : 'codex'
+}
+
+// 按 executor 分组任务
+function groupTasksByExecutor(tasks) {
+  const groups = { gemini: [], codex: [], agent: [] }
+  tasks.forEach(task => {
+    const executor = getTaskExecutor(task)
+    groups[executor].push(task)
+  })
+  return groups
+}
+```
+
+**Execution Flow**: Parallel batches concurrently → Sequential batches in order
+```javascript
+const parallel = executionCalls.filter(c => c.executionType === "parallel")
+const sequential = executionCalls.filter(c => c.executionType === "sequential")
+
+// Phase 1: Launch all parallel batches (single message with multiple tool calls)
+if (parallel.length > 0) {
+  TodoWrite({ todos: executionCalls.map(c => ({ status: c.executionType === "parallel" ? "in_progress" : "pending" })) })
+  parallelResults = await Promise.all(parallel.map(c => executeBatch(c)))
+  previousExecutionResults.push(...parallelResults)
+  TodoWrite({ todos: executionCalls.map(c => ({ status: parallel.includes(c) ? "completed" : "pending" })) })
+}
+
+// Phase 2: Execute sequential batches one by one
+for (const call of sequential) {
+  TodoWrite({ todos: executionCalls.map(c => ({ status: c === call ? "in_progress" : "..." })) })
+  result = await executeBatch(call)
+  previousExecutionResults.push(result)
+  TodoWrite({ todos: executionCalls.map(c => ({ status: "completed" or "pending" })) })
+}
+```
+
+**Option A: Agent Execution**
+
+When to use:
+- `getTaskExecutor(task) === "agent"`
+- 或 `executionMethod = "Agent"` (全局 fallback)
+- 或 `executionMethod = "Auto" AND complexity = "Low"` (全局 fallback)
+
+**Task Formatting Principle**: Each task is a self-contained checklist. The agent only needs to know what THIS task requires, not its position or relation to other tasks.
+
+Agent call format:
+```javascript
+// Format single task as self-contained checklist
+function formatTaskChecklist(task) {
+  return `
+## ${task.title}
+
+**Target**: \`${task.file}\`
+**Action**: ${task.action}
+
+### What to do
+${task.description}
+
+### How to do it
+${task.implementation.map(step => `- ${step}`).join('\n')}
+
+### Reference
+- Pattern: ${task.reference.pattern}
+- Examples: ${task.reference.files.join(', ')}
+- Notes: ${task.reference.examples}
+
+### Done when
+${task.acceptance.map(c => `- [ ] ${c}`).join('\n')}
+`
+}
+
+// For batch execution: aggregate tasks without numbering
+function formatBatchPrompt(batch) {
+  const tasksSection = batch.tasks.map(t => formatTaskChecklist(t)).join('\n---\n')
+
+  return `
+${originalUserInput ? `## Goal\n${originalUserInput}\n` : ''}
+
+## Tasks
+
+${tasksSection}
+
+${batch.context ? `## Context\n${batch.context}` : ''}
+
+Complete each task according to its "Done when" checklist.
+`
+}
+
+Task(
+  subagent_type="code-developer",
+  run_in_background=false,
+  description=batch.taskSummary,
+  prompt=formatBatchPrompt({
+    tasks: batch.tasks,
+    context: buildRelevantContext(batch.tasks)
+  })
+)
+
+// Helper: Build relevant context for batch
+// Context serves as REFERENCE ONLY - helps agent understand existing state
+function buildRelevantContext(tasks) {
+  const sections = []
+
+  // 1. Previous work completion - what's already done (reference for continuity)
+  if (previousExecutionResults.length > 0) {
+    sections.push(`### Previous Work (Reference)
+Use this to understand what's already completed. Avoid duplicating work.
+
+${previousExecutionResults.map(r => `**${r.tasksSummary}**
+- Status: ${r.status}
+- Outputs: ${r.keyOutputs || 'See git diff'}
+${r.notes ? `- Notes: ${r.notes}` : ''}`
+    ).join('\n\n')}`)
+  }
+
+  // 2. Related files - files that may need to be read/referenced
+  const relatedFiles = extractRelatedFiles(tasks)
+  if (relatedFiles.length > 0) {
+    sections.push(`### Related Files (Reference)
+These files may contain patterns, types, or utilities relevant to your tasks:
+
+${relatedFiles.map(f => `- \`${f}\``).join('\n')}`)
+  }
+
+  // 3. Clarifications from user
+  if (clarificationContext) {
+    sections.push(`### User Clarifications
+${Object.entries(clarificationContext).map(([q, a]) => `- **${q}**: ${a}`).join('\n')}`)
+  }
+
+  // 4. Artifact files (for deeper context if needed)
+  if (executionContext?.session?.artifacts?.plan) {
+    sections.push(`### Artifacts
+For detailed planning context, read: ${executionContext.session.artifacts.plan}`)
+  }
+
+  return sections.join('\n\n')
+}
+
+// Extract related files from task references
+function extractRelatedFiles(tasks) {
+  const files = new Set()
+  tasks.forEach(task => {
+    // Add reference example files
+    if (task.reference?.files) {
+      task.reference.files.forEach(f => files.add(f))
+    }
+  })
+  return [...files]
+}
+```
+
+**Result Collection**: After completion, collect result following `executionResult` structure (see Data Structures section)
+
+**Option B: CLI Execution (Codex)**
+
+When to use:
+- `getTaskExecutor(task) === "codex"`
+- 或 `executionMethod = "Codex"` (全局 fallback)
+- 或 `executionMethod = "Auto" AND complexity = "Medium/High"` (全局 fallback)
+
+**Task Formatting Principle**: Same as Agent - each task is a self-contained checklist. No task numbering or position awareness.
+
+Command format:
+```bash
+// Format single task as compact checklist for CLI
+function formatTaskForCLI(task) {
+  return `
+## ${task.title}
+File: ${task.file}
+Action: ${task.action}
+
+What: ${task.description}
+
+How:
+${task.implementation.map(step => `- ${step}`).join('\n')}
+
+Reference: ${task.reference.pattern} (see ${task.reference.files.join(', ')})
+Notes: ${task.reference.examples}
+
+Done when:
+${task.acceptance.map(c => `- [ ] ${c}`).join('\n')}
+`
+}
+
+// Build CLI prompt for batch
+// Context provides REFERENCE information - not requirements to fulfill
+function buildCLIPrompt(batch) {
+  const tasksSection = batch.tasks.map(t => formatTaskForCLI(t)).join('\n---\n')
+
+  let prompt = `${originalUserInput ? `## Goal\n${originalUserInput}\n\n` : ''}`
+  prompt += `## Tasks\n\n${tasksSection}\n`
+
+  // Context section - reference information only
+  const contextSections = []
+
+  // 1. Previous work - what's already completed
+  if (previousExecutionResults.length > 0) {
+    contextSections.push(`### Previous Work (Reference)
+Already completed - avoid duplicating:
+${previousExecutionResults.map(r => `- ${r.tasksSummary}: ${r.status}${r.keyOutputs ? ` (${r.keyOutputs})` : ''}`).join('\n')}`)
+  }
+
+  // 2. Related files from task references
+  const relatedFiles = [...new Set(batch.tasks.flatMap(t => t.reference?.files || []))]
+  if (relatedFiles.length > 0) {
+    contextSections.push(`### Related Files (Reference)
+Patterns and examples to follow:
+${relatedFiles.map(f => `- ${f}`).join('\n')}`)
+  }
+
+  // 3. User clarifications
+  if (clarificationContext) {
+    contextSections.push(`### Clarifications
+${Object.entries(clarificationContext).map(([q, a]) => `- ${q}: ${a}`).join('\n')}`)
+  }
+
+  // 4. Plan artifact for deeper context
+  if (executionContext?.session?.artifacts?.plan) {
+    contextSections.push(`### Artifacts
+Detailed plan: ${executionContext.session.artifacts.plan}`)
+  }
+
+  if (contextSections.length > 0) {
+    prompt += `\n## Context\n${contextSections.join('\n\n')}\n`
+  }
+
+  prompt += `\nComplete each task according to its "Done when" checklist.`
+
+  return prompt
+}
+
+ccw cli -p "${buildCLIPrompt(batch)}" --tool codex --mode write
+```
+
+**Execution with fixed IDs** (predictable ID pattern):
+```javascript
+// Launch CLI in foreground (NOT background)
+// Timeout based on complexity: Low=40min, Medium=60min, High=100min
+const timeoutByComplexity = {
+  "Low": 2400000,    // 40 minutes
+  "Medium": 3600000, // 60 minutes
+  "High": 6000000    // 100 minutes
+}
+
+// Generate fixed execution ID: ${sessionId}-${groupId}
+// This enables predictable ID lookup without relying on resume context chains
+const sessionId = executionContext?.session?.id || 'standalone'
+const fixedExecutionId = `${sessionId}-${batch.groupId}`  // e.g., "implement-auth-2025-12-13-P1"
+
+// Check if resuming from previous failed execution
+const previousCliId = batch.resumeFromCliId || null
+
+// Build command with fixed ID (and optional resume for continuation)
+const cli_command = previousCliId
+  ? `ccw cli -p "${buildCLIPrompt(batch)}" --tool codex --mode write --id ${fixedExecutionId} --resume ${previousCliId}`
+  : `ccw cli -p "${buildCLIPrompt(batch)}" --tool codex --mode write --id ${fixedExecutionId}`
+
+bash_result = Bash(
+  command=cli_command,
+  timeout=timeoutByComplexity[planObject.complexity] || 3600000
+)
+
+// Execution ID is now predictable: ${fixedExecutionId}
+// Can also extract from output: "ID: implement-auth-2025-12-13-P1"
+const cliExecutionId = fixedExecutionId
+
+// Update TodoWrite when execution completes
+```
+
+**Resume on Failure** (with fixed ID):
+```javascript
+// If execution failed or timed out, offer resume option
+if (bash_result.status === 'failed' || bash_result.status === 'timeout') {
+  console.log(`
+⚠️ Execution incomplete. Resume available:
+   Fixed ID: ${fixedExecutionId}
+   Lookup: ccw cli detail ${fixedExecutionId}
+   Resume: ccw cli -p "Continue tasks" --resume ${fixedExecutionId} --tool codex --mode write --id ${fixedExecutionId}-retry
+`)
+
+  // Store for potential retry in same session
+  batch.resumeFromCliId = fixedExecutionId
+}
+```
+
+**Result Collection**: After completion, analyze output and collect result following `executionResult` structure (include `cliExecutionId` for resume capability)
+
+**Option C: CLI Execution (Gemini)**
+
+When to use: `getTaskExecutor(task) === "gemini"` (分析类任务)
+
+```bash
+# 使用与 Option B 相同的 formatBatchPrompt，切换 tool 和 mode
+ccw cli -p "${formatBatchPrompt(batch)}" --tool gemini --mode analysis --id ${sessionId}-${batch.groupId}
+```
+
+### Step 4: Progress Tracking
+
+Progress tracked at batch level (not individual task level). Icons: ⚡ (parallel, concurrent), → (sequential, one-by-one)
+
+### Step 5: Code Review (Optional)
+
+**Skip Condition**: Only run if `codeReviewTool ≠ "Skip"`
+
+**Review Focus**: Verify implementation against plan acceptance criteria
+- Read plan.json for task acceptance criteria
+- Check each acceptance criterion is fulfilled
+- Validate code quality and identify issues
+- Ensure alignment with planned approach
+
+**Operations**:
+- Agent Review: Current agent performs direct review
+- Gemini Review: Execute gemini CLI with review prompt
+- Custom tool: Execute specified CLI tool (qwen, codex, etc.)
+
+**Unified Review Template** (All tools use same standard):
+
+**Review Criteria**:
+- **Acceptance Criteria**: Verify each criterion from plan.tasks[].acceptance
+- **Code Quality**: Analyze quality, identify issues, suggest improvements
+- **Plan Alignment**: Validate implementation matches planned approach
+
+**Shared Prompt Template** (used by all CLI tools):
+```
+PURPOSE: Code review for implemented changes against plan acceptance criteria
+TASK: • Verify plan acceptance criteria fulfillment • Analyze code quality • Identify issues • Suggest improvements • Validate plan adherence
+MODE: analysis
+CONTEXT: @**/* @{plan.json} [@{exploration.json}] | Memory: Review lite-execute changes against plan requirements
+EXPECTED: Quality assessment with acceptance criteria verification, issue identification, and recommendations. Explicitly check each acceptance criterion from plan.json tasks.
+RULES: $(cat ~/.claude/workflows/cli-templates/prompts/analysis/02-review-code-quality.txt) | Focus on plan acceptance criteria and plan adherence | analysis=READ-ONLY
+```
+
+**Tool-Specific Execution** (Apply shared prompt template above):
+
+```bash
+# Method 1: Agent Review (current agent)
+# - Read plan.json: ${executionContext.session.artifacts.plan}
+# - Apply unified review criteria (see Shared Prompt Template)
+# - Report findings directly
+
+# Method 2: Gemini Review (recommended)
+ccw cli -p "[Shared Prompt Template with artifacts]" --tool gemini --mode analysis
+# CONTEXT includes: @**/* @${plan.json} [@${exploration.json}]
+
+# Method 3: Qwen Review (alternative)
+ccw cli -p "[Shared Prompt Template with artifacts]" --tool qwen --mode analysis
+# Same prompt as Gemini, different execution engine
+
+# Method 4: Codex Review (autonomous)
+ccw cli -p "[Verify plan acceptance criteria at ${plan.json}]" --tool codex --mode write
+```
+
+**Multi-Round Review with Fixed IDs**:
+```javascript
+// Generate fixed review ID
+const reviewId = `${sessionId}-review`
+
+// First review pass with fixed ID
+const reviewResult = Bash(`ccw cli -p "[Review prompt]" --tool gemini --mode analysis --id ${reviewId}`)
+
+// If issues found, continue review dialog with fixed ID chain
+if (hasUnresolvedIssues(reviewResult)) {
+  // Resume with follow-up questions
+  Bash(`ccw cli -p "Clarify the security concerns you mentioned" --resume ${reviewId} --tool gemini --mode analysis --id ${reviewId}-followup`)
+}
+```
+
+**Implementation Note**: Replace `[Shared Prompt Template with artifacts]` placeholder with actual template content, substituting:
+- `@{plan.json}` → `@${executionContext.session.artifacts.plan}`
+- `[@{exploration.json}]` → exploration files from artifacts (if exists)
+
+### Step 6: Update Development Index
+
+**Trigger**: After all executions complete (regardless of code review)
+
+**Skip Condition**: Skip if `.workflow/project.json` does not exist
+
+**Operations**:
+```javascript
+const projectJsonPath = '.workflow/project.json'
+if (!fileExists(projectJsonPath)) return  // Silent skip
+
+const projectJson = JSON.parse(Read(projectJsonPath))
+
+// Initialize if needed
+if (!projectJson.development_index) {
+  projectJson.development_index = { feature: [], enhancement: [], bugfix: [], refactor: [], docs: [] }
+}
+
+// Detect category from keywords
+function detectCategory(text) {
+  text = text.toLowerCase()
+  if (/\b(fix|bug|error|issue|crash)\b/.test(text)) return 'bugfix'
+  if (/\b(refactor|cleanup|reorganize)\b/.test(text)) return 'refactor'
+  if (/\b(doc|readme|comment)\b/.test(text)) return 'docs'
+  if (/\b(add|new|create|implement)\b/.test(text)) return 'feature'
+  return 'enhancement'
+}
+
+// Detect sub_feature from task file paths
+function detectSubFeature(tasks) {
+  const dirs = tasks.map(t => t.file?.split('/').slice(-2, -1)[0]).filter(Boolean)
+  const counts = dirs.reduce((a, d) => { a[d] = (a[d] || 0) + 1; return a }, {})
+  return Object.entries(counts).sort((a, b) => b[1] - a[1])[0]?.[0] || 'general'
+}
+
+const category = detectCategory(`${planObject.summary} ${planObject.approach}`)
+const entry = {
+  title: planObject.summary.slice(0, 60),
+  sub_feature: detectSubFeature(planObject.tasks),
+  date: new Date().toISOString().split('T')[0],
+  description: planObject.approach.slice(0, 100),
+  status: previousExecutionResults.every(r => r.status === 'completed') ? 'completed' : 'partial',
+  session_id: executionContext?.session?.id || null
+}
+
+projectJson.development_index[category].push(entry)
+projectJson.statistics.last_updated = new Date().toISOString()
+Write(projectJsonPath, JSON.stringify(projectJson, null, 2))
+
+console.log(`✓ Development index: [${category}] ${entry.title}`)
+```
+
+## Best Practices
+
+**Input Modes**: In-memory (lite-plan), prompt (standalone), file (JSON/text)
+**Task Grouping**: Based on explicit depends_on only; independent tasks run in single parallel batch
+**Execution**: All independent tasks launch concurrently via single Claude message with multiple tool calls
+
+## Error Handling
+
+| Error | Cause | Resolution |
+|-------|-------|------------|
+| Missing executionContext | --in-memory without context | Error: "No execution context found. Only available when called by lite-plan." |
+| File not found | File path doesn't exist | Error: "File not found: {path}. Check file path." |
+| Empty file | File exists but no content | Error: "File is empty: {path}. Provide task description." |
+| Invalid Enhanced Task JSON | JSON missing required fields | Warning: "Missing required fields. Treating as plain text." |
+| Malformed JSON | JSON parsing fails | Treat as plain text (expected for non-JSON files) |
+| Execution failure | Agent/Codex crashes | Display error, use fixed ID `${sessionId}-${groupId}` for resume: `ccw cli -p "Continue" --resume <fixed-id> --id <fixed-id>-retry` |
+| Execution timeout | CLI exceeded timeout | Use fixed ID for resume with extended timeout |
+| Codex unavailable | Codex not installed | Show installation instructions, offer Agent execution |
+| Fixed ID not found | Custom ID lookup failed | Check `ccw cli history`, verify date directories |
+
+## Data Structures
+
+### executionContext (Input - Mode 1)
+
+Passed from lite-plan via global variable:
+
+```javascript
+{
+  planObject: {
+    summary: string,
+    approach: string,
+    tasks: [...],
+    estimated_time: string,
+    recommended_execution: string,
+    complexity: string
+  },
+  explorationsContext: {...} | null,       // Multi-angle explorations
+  explorationAngles: string[],             // List of exploration angles
+  explorationManifest: {...} | null,       // Exploration manifest
+  clarificationContext: {...} | null,
+  executionMethod: "Agent" | "Codex" | "Auto",  // 全局默认
+  codeReviewTool: "Skip" | "Gemini Review" | "Agent Review" | string,
+  originalUserInput: string,
+
+  // 任务级 executor 分配（优先于 executionMethod）
+  executorAssignments: {
+    [taskId]: { executor: "gemini" | "codex" | "agent", reason: string }
+  },
+
+  // Session artifacts location (saved by lite-plan)
+  session: {
+    id: string,                        // Session identifier: {taskSlug}-{shortTimestamp}
+    folder: string,                    // Session folder path: .workflow/.lite-plan/{session-id}
+    artifacts: {
+      explorations: [{angle, path}],   // exploration-{angle}.json paths
+      explorations_manifest: string,   // explorations-manifest.json path
+      plan: string                     // plan.json path (always present)
+    }
+  }
+}
+```
+
+**Artifact Usage**:
+- Artifact files contain detailed planning context
+- Pass artifact paths to CLI tools and agents for enhanced context
+- See execution options below for usage examples
+
+### executionResult (Output)
+
+Collected after each execution call completes:
+
+```javascript
+{
+  executionId: string,                 // e.g., "[Agent-1]", "[Codex-1]"
+  status: "completed" | "partial" | "failed",
+  tasksSummary: string,                // Brief description of tasks handled
+  completionSummary: string,           // What was completed
+  keyOutputs: string,                  // Files created/modified, key changes
+  notes: string,                       // Important context for next execution
+  fixedCliId: string | null            // Fixed CLI execution ID (e.g., "implement-auth-2025-12-13-P1")
+}
+```
+
+Appended to `previousExecutionResults` array for context continuity in multi-execution scenarios.
+
+**Fixed ID Pattern**: `${sessionId}-${groupId}` enables predictable lookup without auto-generated timestamps.
+
+**Resume Usage**: If `status` is "partial" or "failed", use `fixedCliId` to resume:
+```bash
+# Lookup previous execution
+ccw cli detail ${fixedCliId}
+
+# Resume with new fixed ID for retry
+ccw cli -p "Continue from where we left off" --resume ${fixedCliId} --tool codex --mode write --id ${fixedCliId}-retry
+```

.claude/commands/workflow/lite-fix.md

@@ -0,0 +1,631 @@
+---
+name: lite-fix
+description: Lightweight bug diagnosis and fix workflow with intelligent severity assessment and optional hotfix mode for production incidents
+argument-hint: "[--hotfix] \"bug description or issue reference\""
+allowed-tools: TodoWrite(*), Task(*), SlashCommand(*), AskUserQuestion(*)
+---
+
+# Workflow Lite-Fix Command (/workflow:lite-fix)
+
+## Overview
+
+Intelligent lightweight bug fixing command with dynamic workflow adaptation based on severity assessment. Focuses on diagnosis phases (root cause analysis, impact assessment, fix planning, confirmation) and delegates execution to `/workflow:lite-execute`.
+
+**Core capabilities:**
+- Intelligent bug analysis with automatic severity detection
+- Dynamic code diagnosis (cli-explore-agent) for root cause identification
+- Interactive clarification after diagnosis to gather missing information
+- Adaptive fix planning strategy (direct Claude vs cli-lite-planning-agent) based on complexity
+- Two-step confirmation: fix-plan display -> multi-dimensional input collection
+- Execution dispatch with complete context handoff to lite-execute
+
+## Usage
+
+```bash
+/workflow:lite-fix [FLAGS] <BUG_DESCRIPTION>
+
+# Flags
+--hotfix, -h               Production hotfix mode (minimal diagnosis, fast fix)
+
+# Arguments
+<bug-description>          Bug description, error message, or path to .md file (required)
+```
+
+## Execution Process
+
+```
+Phase 1: Bug Analysis & Diagnosis
+   |- Parse input (description, error message, or .md file)
+   |- Intelligent severity pre-assessment (Low/Medium/High/Critical)
+   |- Diagnosis decision (auto-detect or --hotfix flag)
+   |- Context protection: If file reading >=50k chars -> force cli-explore-agent
+   +- Decision:
+      |- needsDiagnosis=true -> Launch parallel cli-explore-agents (1-4 based on severity)
+      +- needsDiagnosis=false (hotfix) -> Skip directly to Phase 3 (Fix Planning)
+
+Phase 2: Clarification (optional, multi-round)
+   |- Aggregate clarification_needs from all diagnosis angles
+   |- Deduplicate similar questions
+   +- Decision:
+      |- Has clarifications -> AskUserQuestion (max 4 questions per round, multiple rounds allowed)
+      +- No clarifications -> Skip to Phase 3
+
+Phase 3: Fix Planning (NO CODE EXECUTION - planning only)
+   +- Decision (based on Phase 1 severity):
+      |- Low/Medium -> Load schema: cat ~/.claude/workflows/cli-templates/schemas/fix-plan-json-schema.json -> Direct Claude planning (following schema) -> fix-plan.json -> MUST proceed to Phase 4
+      +- High/Critical -> cli-lite-planning-agent -> fix-plan.json -> MUST proceed to Phase 4
+
+Phase 4: Confirmation & Selection
+   |- Display fix-plan summary (tasks, severity, estimated time)
+   +- AskUserQuestion:
+      |- Confirm: Allow / Modify / Cancel
+      |- Execution: Agent / Codex / Auto
+      +- Review: Gemini / Agent / Skip
+
+Phase 5: Dispatch
+   |- Build executionContext (fix-plan + diagnoses + clarifications + selections)
+   +- SlashCommand("/workflow:lite-execute --in-memory --mode bugfix")
+```
+
+## Implementation
+
+### Phase 1: Intelligent Multi-Angle Diagnosis
+
+**Session Setup** (MANDATORY - follow exactly):
+```javascript
+// Helper: Get UTC+8 (China Standard Time) ISO string
+const getUtc8ISOString = () => new Date(Date.now() + 8 * 60 * 60 * 1000).toISOString()
+
+const bugSlug = bug_description.toLowerCase().replace(/[^a-z0-9]+/g, '-').substring(0, 40)
+const dateStr = getUtc8ISOString().substring(0, 10)  // Format: 2025-11-29
+
+const sessionId = `${bugSlug}-${dateStr}`  // e.g., "user-avatar-upload-fails-2025-11-29"
+const sessionFolder = `.workflow/.lite-fix/${sessionId}`
+
+bash(`mkdir -p ${sessionFolder} && test -d ${sessionFolder} && echo "SUCCESS: ${sessionFolder}" || echo "FAILED: ${sessionFolder}"`)
+```
+
+**Diagnosis Decision Logic**:
+```javascript
+const hotfixMode = $ARGUMENTS.includes('--hotfix') || $ARGUMENTS.includes('-h')
+
+needsDiagnosis = (
+  !hotfixMode &&
+  (
+    bug.lacks_specific_error_message ||
+    bug.requires_codebase_context ||
+    bug.needs_execution_tracing ||
+    bug.root_cause_unclear
+  )
+)
+
+if (!needsDiagnosis) {
+  // Skip to Phase 2 (Clarification) or Phase 3 (Fix Planning)
+  proceed_to_next_phase()
+}
+```
+
+**Context Protection**: File reading >=50k chars -> force `needsDiagnosis=true` (delegate to cli-explore-agent)
+
+**Severity Pre-Assessment** (Intelligent Analysis):
+```javascript
+// Analyzes bug severity based on:
+// - Symptoms: Error messages, crash reports, user complaints
+// - Scope: How many users/features are affected?
+// - Urgency: Production down vs minor inconvenience
+// - Impact: Data loss, security, business impact
+
+const severity = analyzeBugSeverity(bug_description)
+// Returns: 'Low' | 'Medium' | 'High' | 'Critical'
+// Low: Minor UI issue, localized, no data impact
+// Medium: Multiple users affected, degraded functionality
+// High: Significant functionality broken, many users affected
+// Critical: Production down, data loss risk, security issue
+
+// Angle assignment based on bug type (orchestrator decides, not agent)
+const DIAGNOSIS_ANGLE_PRESETS = {
+  runtime_error: ['error-handling', 'dataflow', 'state-management', 'edge-cases'],
+  performance: ['performance', 'bottlenecks', 'caching', 'data-access'],
+  security: ['security', 'auth-patterns', 'dataflow', 'validation'],
+  data_corruption: ['data-integrity', 'state-management', 'transactions', 'validation'],
+  ui_bug: ['state-management', 'event-handling', 'rendering', 'data-binding'],
+  integration: ['api-contracts', 'error-handling', 'timeouts', 'fallbacks']
+}
+
+function selectDiagnosisAngles(bugDescription, count) {
+  const text = bugDescription.toLowerCase()
+  let preset = 'runtime_error' // default
+
+  if (/slow|timeout|performance|lag|hang/.test(text)) preset = 'performance'
+  else if (/security|auth|permission|access|token/.test(text)) preset = 'security'
+  else if (/corrupt|data|lost|missing|inconsistent/.test(text)) preset = 'data_corruption'
+  else if (/ui|display|render|style|click|button/.test(text)) preset = 'ui_bug'
+  else if (/api|integration|connect|request|response/.test(text)) preset = 'integration'
+
+  return DIAGNOSIS_ANGLE_PRESETS[preset].slice(0, count)
+}
+
+const selectedAngles = selectDiagnosisAngles(bug_description, severity === 'Critical' ? 4 : (severity === 'High' ? 3 : (severity === 'Medium' ? 2 : 1)))
+
+console.log(`
+## Diagnosis Plan
+
+Bug Severity: ${severity}
+Selected Angles: ${selectedAngles.join(', ')}
+
+Launching ${selectedAngles.length} parallel diagnoses...
+`)
+```
+
+**Launch Parallel Diagnoses** - Orchestrator assigns angle to each agent:
+
+```javascript
+// Launch agents with pre-assigned diagnosis angles
+const diagnosisTasks = selectedAngles.map((angle, index) =>
+  Task(
+    subagent_type="cli-explore-agent",
+    run_in_background=false,
+    description=`Diagnose: ${angle}`,
+    prompt=`
+## Task Objective
+Execute **${angle}** diagnosis for bug root cause analysis. Analyze codebase from this specific angle to discover root cause, affected paths, and fix hints.
+
+## Assigned Context
+- **Diagnosis Angle**: ${angle}
+- **Bug Description**: ${bug_description}
+- **Diagnosis Index**: ${index + 1} of ${selectedAngles.length}
+- **Output File**: ${sessionFolder}/diagnosis-${angle}.json
+
+## MANDATORY FIRST STEPS (Execute by Agent)
+**You (cli-explore-agent) MUST execute these steps in order:**
+1. Run: ccw tool exec get_modules_by_depth '{}' (project structure)
+2. Run: rg -l "{error_keyword_from_bug}" --type ts (locate relevant files)
+3. Execute: cat ~/.claude/workflows/cli-templates/schemas/diagnosis-json-schema.json (get output schema reference)
+4. Read: .workflow/project-tech.json (technology stack and architecture context)
+5. Read: .workflow/project-guidelines.json (user-defined constraints and conventions)
+
+## Diagnosis Strategy (${angle} focus)
+
+**Step 1: Error Tracing** (Bash)
+- rg for error messages, stack traces, log patterns
+- git log --since='2 weeks ago' for recent changes
+- Trace execution path in affected modules
+
+**Step 2: Root Cause Analysis** (Gemini CLI)
+- What code paths lead to this ${angle} issue?
+- What edge cases are not handled from ${angle} perspective?
+- What recent changes might have introduced this bug?
+
+**Step 3: Write Output**
+- Consolidate ${angle} findings into JSON
+- Identify ${angle}-specific clarification needs
+- Provide fix hints based on ${angle} analysis
+
+## Expected Output
+
+**File**: ${sessionFolder}/diagnosis-${angle}.json
+
+**Schema Reference**: Schema obtained in MANDATORY FIRST STEPS step 3, follow schema exactly
+
+**Required Fields** (all ${angle} focused):
+- symptom: Bug symptoms and error messages
+- root_cause: Root cause hypothesis from ${angle} perspective
+  **IMPORTANT**: Use structured format:
+  \`{file: "src/module/file.ts", line_range: "45-60", issue: "Description", confidence: 0.85}\`
+- affected_files: Files involved from ${angle} perspective
+  **IMPORTANT**: Use object format with relevance scores:
+  \`[{path: "src/file.ts", relevance: 0.85, rationale: "Contains ${angle} logic"}]\`
+- reproduction_steps: Steps to reproduce the bug
+- fix_hints: Suggested fix approaches from ${angle} viewpoint
+- dependencies: Dependencies relevant to ${angle} diagnosis
+- constraints: ${angle}-specific limitations affecting fix
+- clarification_needs: ${angle}-related ambiguities (options array + recommended index)
+- _metadata.diagnosis_angle: "${angle}"
+- _metadata.diagnosis_index: ${index + 1}
+
+## Success Criteria
+- [ ] Schema obtained via cat diagnosis-json-schema.json
+- [ ] get_modules_by_depth.sh executed
+- [ ] Root cause identified with confidence score
+- [ ] At least 3 affected files identified with ${angle} rationale
+- [ ] Fix hints are actionable (specific code changes, not generic advice)
+- [ ] Reproduction steps are verifiable
+- [ ] JSON output follows schema exactly
+- [ ] clarification_needs includes options + recommended
+
+## Output
+Write: ${sessionFolder}/diagnosis-${angle}.json
+Return: 2-3 sentence summary of ${angle} diagnosis findings
+`
+  )
+)
+
+// Execute all diagnosis tasks in parallel
+```
+
+**Auto-discover Generated Diagnosis Files**:
+```javascript
+// After diagnoses complete, auto-discover all diagnosis-*.json files
+const diagnosisFiles = bash(`find ${sessionFolder} -name "diagnosis-*.json" -type f`)
+  .split('\n')
+  .filter(f => f.trim())
+
+// Read metadata to build manifest
+const diagnosisManifest = {
+  session_id: sessionId,
+  bug_description: bug_description,
+  timestamp: getUtc8ISOString(),
+  severity: severity,
+  diagnosis_count: diagnosisFiles.length,
+  diagnoses: diagnosisFiles.map(file => {
+    const data = JSON.parse(Read(file))
+    const filename = path.basename(file)
+    return {
+      angle: data._metadata.diagnosis_angle,
+      file: filename,
+      path: file,
+      index: data._metadata.diagnosis_index
+    }
+  })
+}
+
+Write(`${sessionFolder}/diagnoses-manifest.json`, JSON.stringify(diagnosisManifest, null, 2))
+
+console.log(`
+## Diagnosis Complete
+
+Generated diagnosis files in ${sessionFolder}:
+${diagnosisManifest.diagnoses.map(d => `- diagnosis-${d.angle}.json (angle: ${d.angle})`).join('\n')}
+
+Manifest: diagnoses-manifest.json
+Angles diagnosed: ${diagnosisManifest.diagnoses.map(d => d.angle).join(', ')}
+`)
+```
+
+**Output**:
+- `${sessionFolder}/diagnosis-{angle1}.json`
+- `${sessionFolder}/diagnosis-{angle2}.json`
+- ... (1-4 files based on severity)
+- `${sessionFolder}/diagnoses-manifest.json`
+
+---
+
+### Phase 2: Clarification (Optional, Multi-Round)
+
+**Skip if**: No diagnosis or `clarification_needs` is empty across all diagnoses
+
+**⚠️ CRITICAL**: AskUserQuestion tool limits max 4 questions per call. **MUST execute multiple rounds** to exhaust all clarification needs - do NOT stop at round 1.
+
+**Aggregate clarification needs from all diagnosis angles**:
+```javascript
+// Load manifest and all diagnosis files
+const manifest = JSON.parse(Read(`${sessionFolder}/diagnoses-manifest.json`))
+const diagnoses = manifest.diagnoses.map(diag => ({
+  angle: diag.angle,
+  data: JSON.parse(Read(diag.path))
+}))
+
+// Aggregate clarification needs from all diagnoses
+const allClarifications = []
+diagnoses.forEach(diag => {
+  if (diag.data.clarification_needs?.length > 0) {
+    diag.data.clarification_needs.forEach(need => {
+      allClarifications.push({
+        ...need,
+        source_angle: diag.angle
+      })
+    })
+  }
+})
+
+// Deduplicate by question similarity
+function deduplicateClarifications(clarifications) {
+  const unique = []
+  clarifications.forEach(c => {
+    const isDuplicate = unique.some(u =>
+      u.question.toLowerCase() === c.question.toLowerCase()
+    )
+    if (!isDuplicate) unique.push(c)
+  })
+  return unique
+}
+
+const uniqueClarifications = deduplicateClarifications(allClarifications)
+
+// Multi-round clarification: batch questions (max 4 per round)
+// ⚠️ MUST execute ALL rounds until uniqueClarifications exhausted
+if (uniqueClarifications.length > 0) {
+  const BATCH_SIZE = 4
+  const totalRounds = Math.ceil(uniqueClarifications.length / BATCH_SIZE)
+
+  for (let i = 0; i < uniqueClarifications.length; i += BATCH_SIZE) {
+    const batch = uniqueClarifications.slice(i, i + BATCH_SIZE)
+    const currentRound = Math.floor(i / BATCH_SIZE) + 1
+
+    console.log(`### Clarification Round ${currentRound}/${totalRounds}`)
+
+    AskUserQuestion({
+      questions: batch.map(need => ({
+        question: `[${need.source_angle}] ${need.question}\n\nContext: ${need.context}`,
+        header: need.source_angle,
+        multiSelect: false,
+        options: need.options.map((opt, index) => {
+          const isRecommended = need.recommended === index
+          return {
+            label: isRecommended ? `${opt} ★` : opt,
+            description: isRecommended ? `Use ${opt} approach (Recommended)` : `Use ${opt} approach`
+          }
+        })
+      }))
+    })
+
+    // Store batch responses in clarificationContext before next round
+  }
+}
+```
+
+**Output**: `clarificationContext` (in-memory)
+
+---
+
+### Phase 3: Fix Planning
+
+**Planning Strategy Selection** (based on Phase 1 severity):
+
+**IMPORTANT**: Phase 3 is **planning only** - NO code execution. All execution happens in Phase 5 via lite-execute.
+
+**Low/Medium Severity** - Direct planning by Claude:
+```javascript
+// Step 1: Read schema
+const schema = Bash(`cat ~/.claude/workflows/cli-templates/schemas/fix-plan-json-schema.json`)
+
+// Step 2: Generate fix-plan following schema (Claude directly, no agent)
+const fixPlan = {
+  summary: "...",
+  root_cause: "...",
+  strategy: "immediate_patch|comprehensive_fix|refactor",
+  tasks: [...],  // Each task: { id, title, scope, ..., depends_on, complexity }
+  estimated_time: "...",
+  recommended_execution: "Agent",
+  severity: severity,
+  risk_level: "...",
+  _metadata: { timestamp: getUtc8ISOString(), source: "direct-planning", planning_mode: "direct" }
+}
+
+// Step 3: Write fix-plan to session folder
+Write(`${sessionFolder}/fix-plan.json`, JSON.stringify(fixPlan, null, 2))
+
+// Step 4: MUST continue to Phase 4 (Confirmation) - DO NOT execute code here
+```
+
+**High/Critical Severity** - Invoke cli-lite-planning-agent:
+
+```javascript
+Task(
+  subagent_type="cli-lite-planning-agent",
+  run_in_background=false,
+  description="Generate detailed fix plan",
+  prompt=`
+Generate fix plan and write fix-plan.json.
+
+## Output Schema Reference
+Execute: cat ~/.claude/workflows/cli-templates/schemas/fix-plan-json-schema.json (get schema reference before generating plan)
+
+## Project Context (MANDATORY - Read Both Files)
+1. Read: .workflow/project-tech.json (technology stack, architecture, key components)
+2. Read: .workflow/project-guidelines.json (user-defined constraints and conventions)
+
+**CRITICAL**: All fix tasks MUST comply with constraints in project-guidelines.json
+
+## Bug Description
+${bug_description}
+
+## Multi-Angle Diagnosis Context
+
+${manifest.diagnoses.map(diag => `### Diagnosis: ${diag.angle} (${diag.file})
+Path: ${diag.path}
+
+Read this file for detailed ${diag.angle} analysis.`).join('\n\n')}
+
+Total diagnoses: ${manifest.diagnosis_count}
+Angles covered: ${manifest.diagnoses.map(d => d.angle).join(', ')}
+
+Manifest: ${sessionFolder}/diagnoses-manifest.json
+
+## User Clarifications
+${JSON.stringify(clarificationContext) || "None"}
+
+## Severity Level
+${severity}
+
+## Requirements
+Generate fix-plan.json with:
+- summary: 2-3 sentence overview of the fix
+- root_cause: Consolidated root cause from all diagnoses
+- strategy: "immediate_patch" | "comprehensive_fix" | "refactor"
+- tasks: 1-5 structured fix tasks (**IMPORTANT: group by fix area, NOT by file**)
+  - **Task Granularity Principle**: Each task = one complete fix unit
+  - title: action verb + target (e.g., "Fix token validation edge case")
+  - scope: module path (src/auth/) or feature name
+  - action: "Fix" | "Update" | "Refactor" | "Add" | "Delete"
+  - description
+  - modification_points: ALL files to modify for this fix (group related changes)
+  - implementation (2-5 steps covering all modification_points)
+  - verification (test criteria)
+  - depends_on: task IDs this task depends on (use sparingly)
+- estimated_time, recommended_execution, severity, risk_level
+- _metadata:
+  - timestamp, source, planning_mode
+  - diagnosis_angles: ${JSON.stringify(manifest.diagnoses.map(d => d.angle))}
+
+## Task Grouping Rules
+1. **Group by fix area**: All changes for one fix = one task (even if 2-3 files)
+2. **Avoid file-per-task**: Do NOT create separate tasks for each file
+3. **Substantial tasks**: Each task should represent 10-45 minutes of work
+4. **True dependencies only**: Only use depends_on when Task B cannot start without Task A's output
+5. **Prefer parallel**: Most tasks should be independent (no depends_on)
+
+## Execution
+1. Read ALL diagnosis files for comprehensive context
+2. Execute CLI planning using Gemini (Qwen fallback)
+3. Synthesize findings from multiple diagnosis angles
+4. Parse output and structure fix-plan
+5. Write JSON: Write('${sessionFolder}/fix-plan.json', jsonContent)
+6. Return brief completion summary
+`
+)
+```
+
+**Output**: `${sessionFolder}/fix-plan.json`
+
+---
+
+### Phase 4: Task Confirmation & Execution Selection
+
+**Step 4.1: Display Fix Plan**
+```javascript
+const fixPlan = JSON.parse(Read(`${sessionFolder}/fix-plan.json`))
+
+console.log(`
+## Fix Plan
+
+**Summary**: ${fixPlan.summary}
+**Root Cause**: ${fixPlan.root_cause}
+**Strategy**: ${fixPlan.strategy}
+
+**Tasks** (${fixPlan.tasks.length}):
+${fixPlan.tasks.map((t, i) => `${i+1}. ${t.title} (${t.scope})`).join('\n')}
+
+**Severity**: ${fixPlan.severity}
+**Risk Level**: ${fixPlan.risk_level}
+**Estimated Time**: ${fixPlan.estimated_time}
+**Recommended**: ${fixPlan.recommended_execution}
+`)
+```
+
+**Step 4.2: Collect Confirmation**
+```javascript
+AskUserQuestion({
+  questions: [
+    {
+      question: `Confirm fix plan? (${fixPlan.tasks.length} tasks, ${fixPlan.severity} severity)`,
+      header: "Confirm",
+      multiSelect: true,
+      options: [
+        { label: "Allow", description: "Proceed as-is" },
+        { label: "Modify", description: "Adjust before execution" },
+        { label: "Cancel", description: "Abort workflow" }
+      ]
+    },
+    {
+      question: "Execution method:",
+      header: "Execution",
+      multiSelect: false,
+      options: [
+        { label: "Agent", description: "@code-developer agent" },
+        { label: "Codex", description: "codex CLI tool" },
+        { label: "Auto", description: `Auto: ${fixPlan.severity === 'Low' ? 'Agent' : 'Codex'}` }
+      ]
+    },
+    {
+      question: "Code review after fix?",
+      header: "Review",
+      multiSelect: false,
+      options: [
+        { label: "Gemini Review", description: "Gemini CLI" },
+        { label: "Agent Review", description: "@code-reviewer" },
+        { label: "Skip", description: "No review" }
+      ]
+    }
+  ]
+})
+```
+
+---
+
+### Phase 5: Dispatch to Execution
+
+**CRITICAL**: lite-fix NEVER executes code directly. ALL execution MUST go through lite-execute.
+
+**Step 5.1: Build executionContext**
+
+```javascript
+// Load manifest and all diagnosis files
+const manifest = JSON.parse(Read(`${sessionFolder}/diagnoses-manifest.json`))
+const diagnoses = {}
+
+manifest.diagnoses.forEach(diag => {
+  if (file_exists(diag.path)) {
+    diagnoses[diag.angle] = JSON.parse(Read(diag.path))
+  }
+})
+
+const fixPlan = JSON.parse(Read(`${sessionFolder}/fix-plan.json`))
+
+executionContext = {
+  mode: "bugfix",
+  severity: fixPlan.severity,
+  planObject: fixPlan,
+  diagnosisContext: diagnoses,
+  diagnosisAngles: manifest.diagnoses.map(d => d.angle),
+  diagnosisManifest: manifest,
+  clarificationContext: clarificationContext || null,
+  executionMethod: userSelection.execution_method,
+  codeReviewTool: userSelection.code_review_tool,
+  originalUserInput: bug_description,
+  session: {
+    id: sessionId,
+    folder: sessionFolder,
+    artifacts: {
+      diagnoses: manifest.diagnoses.map(diag => ({
+        angle: diag.angle,
+        path: diag.path
+      })),
+      diagnoses_manifest: `${sessionFolder}/diagnoses-manifest.json`,
+      fix_plan: `${sessionFolder}/fix-plan.json`
+    }
+  }
+}
+```
+
+**Step 5.2: Dispatch**
+
+```javascript
+SlashCommand(command="/workflow:lite-execute --in-memory --mode bugfix")
+```
+
+## Session Folder Structure
+
+```
+.workflow/.lite-fix/{bug-slug}-{YYYY-MM-DD}/
+|- diagnosis-{angle1}.json      # Diagnosis angle 1
+|- diagnosis-{angle2}.json      # Diagnosis angle 2
+|- diagnosis-{angle3}.json      # Diagnosis angle 3 (if applicable)
+|- diagnosis-{angle4}.json      # Diagnosis angle 4 (if applicable)
+|- diagnoses-manifest.json      # Diagnosis index
++- fix-plan.json                # Fix plan
+```
+
+**Example**:
+```
+.workflow/.lite-fix/user-avatar-upload-fails-413-2025-11-25-14-30-25/
+|- diagnosis-error-handling.json
+|- diagnosis-dataflow.json
+|- diagnosis-validation.json
+|- diagnoses-manifest.json
++- fix-plan.json
+```
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| Diagnosis agent failure | Skip diagnosis, continue with bug description only |
+| Planning agent failure | Fallback to direct planning by Claude |
+| Clarification timeout | Use diagnosis findings as-is |
+| Confirmation timeout | Save context, display resume instructions |
+| Modify loop > 3 times | Suggest breaking task or using /workflow:plan |
+| Root cause unclear | Extend diagnosis time or use broader angles |
+| Too complex for lite-fix | Escalate to /workflow:plan --mode bugfix |
+
+

.claude/commands/workflow/lite-plan.md

@@ -0,0 +1,621 @@
+---
+name: lite-plan
+description: Lightweight interactive planning workflow with in-memory planning, code exploration, and execution dispatch to lite-execute after user confirmation
+argument-hint: "[-e|--explore] \"task description\"|file.md"
+allowed-tools: TodoWrite(*), Task(*), SlashCommand(*), AskUserQuestion(*)
+---
+
+# Workflow Lite-Plan Command (/workflow:lite-plan)
+
+## Overview
+
+Intelligent lightweight planning command with dynamic workflow adaptation based on task complexity. Focuses on planning phases (exploration, clarification, planning, confirmation) and delegates execution to `/workflow:lite-execute`.
+
+**Core capabilities:**
+- Intelligent task analysis with automatic exploration detection
+- Dynamic code exploration (cli-explore-agent) when codebase understanding needed
+- Interactive clarification after exploration to gather missing information
+- Adaptive planning: Low complexity → Direct Claude; Medium/High → cli-lite-planning-agent
+- Two-step confirmation: plan display → multi-dimensional input collection
+- Execution dispatch with complete context handoff to lite-execute
+
+## Usage
+
+```bash
+/workflow:lite-plan [FLAGS] <TASK_DESCRIPTION>
+
+# Flags
+-e, --explore              Force code exploration phase (overrides auto-detection)
+
+# Arguments
+<task-description>         Task description or path to .md file (required)
+```
+
+## Execution Process
+
+```
+Phase 1: Task Analysis & Exploration
+   ├─ Parse input (description or .md file)
+   ├─ intelligent complexity assessment (Low/Medium/High)
+   ├─ Exploration decision (auto-detect or --explore flag)
+   ├─ Context protection: If file reading ≥50k chars → force cli-explore-agent
+   └─ Decision:
+      ├─ needsExploration=true → Launch parallel cli-explore-agents (1-4 based on complexity)
+      └─ needsExploration=false → Skip to Phase 2/3
+
+Phase 2: Clarification (optional, multi-round)
+   ├─ Aggregate clarification_needs from all exploration angles
+   ├─ Deduplicate similar questions
+   └─ Decision:
+      ├─ Has clarifications → AskUserQuestion (max 4 questions per round, multiple rounds allowed)
+      └─ No clarifications → Skip to Phase 3
+
+Phase 3: Planning (NO CODE EXECUTION - planning only)
+   └─ Decision (based on Phase 1 complexity):
+      ├─ Low → Load schema: cat ~/.claude/workflows/cli-templates/schemas/plan-json-schema.json → Direct Claude planning (following schema) → plan.json → MUST proceed to Phase 4
+      └─ Medium/High → cli-lite-planning-agent → plan.json → MUST proceed to Phase 4
+
+Phase 4: Confirmation & Selection
+   ├─ Display plan summary (tasks, complexity, estimated time)
+   └─ AskUserQuestion:
+      ├─ Confirm: Allow / Modify / Cancel
+      ├─ Execution: Agent / Codex / Auto
+      └─ Review: Gemini / Agent / Skip
+
+Phase 5: Dispatch
+   ├─ Build executionContext (plan + explorations + clarifications + selections)
+   └─ SlashCommand("/workflow:lite-execute --in-memory")
+```
+
+## Implementation
+
+### Phase 1: Intelligent Multi-Angle Exploration
+
+**Session Setup** (MANDATORY - follow exactly):
+```javascript
+// Helper: Get UTC+8 (China Standard Time) ISO string
+const getUtc8ISOString = () => new Date(Date.now() + 8 * 60 * 60 * 1000).toISOString()
+
+const taskSlug = task_description.toLowerCase().replace(/[^a-z0-9]+/g, '-').substring(0, 40)
+const dateStr = getUtc8ISOString().substring(0, 10)  // Format: 2025-11-29
+
+const sessionId = `${taskSlug}-${dateStr}`  // e.g., "implement-jwt-refresh-2025-11-29"
+const sessionFolder = `.workflow/.lite-plan/${sessionId}`
+
+bash(`mkdir -p ${sessionFolder} && test -d ${sessionFolder} && echo "SUCCESS: ${sessionFolder}" || echo "FAILED: ${sessionFolder}"`)
+```
+
+**Exploration Decision Logic**:
+```javascript
+needsExploration = (
+  flags.includes('--explore') || flags.includes('-e') ||
+  task.mentions_specific_files ||
+  task.requires_codebase_context ||
+  task.needs_architecture_understanding ||
+  task.modifies_existing_code
+)
+
+if (!needsExploration) {
+  // Skip to Phase 2 (Clarification) or Phase 3 (Planning)
+  proceed_to_next_phase()
+}
+```
+
+**⚠️ Context Protection**: File reading ≥50k chars → force `needsExploration=true` (delegate to cli-explore-agent)
+
+**Complexity Assessment** (Intelligent Analysis):
+```javascript
+// analyzes task complexity based on:
+// - Scope: How many systems/modules are affected?
+// - Depth: Surface change vs architectural impact?
+// - Risk: Potential for breaking existing functionality?
+// - Dependencies: How interconnected is the change?
+
+const complexity = analyzeTaskComplexity(task_description)
+// Returns: 'Low' | 'Medium' | 'High'
+// Low: Single file, isolated change, minimal risk
+// Medium: Multiple files, some dependencies, moderate risk
+// High: Cross-module, architectural, high risk
+
+// Angle assignment based on task type (orchestrator decides, not agent)
+const ANGLE_PRESETS = {
+  architecture: ['architecture', 'dependencies', 'modularity', 'integration-points'],
+  security: ['security', 'auth-patterns', 'dataflow', 'validation'],
+  performance: ['performance', 'bottlenecks', 'caching', 'data-access'],
+  bugfix: ['error-handling', 'dataflow', 'state-management', 'edge-cases'],
+  feature: ['patterns', 'integration-points', 'testing', 'dependencies']
+}
+
+function selectAngles(taskDescription, count) {
+  const text = taskDescription.toLowerCase()
+  let preset = 'feature' // default
+
+  if (/refactor|architect|restructure|modular/.test(text)) preset = 'architecture'
+  else if (/security|auth|permission|access/.test(text)) preset = 'security'
+  else if (/performance|slow|optimi|cache/.test(text)) preset = 'performance'
+  else if (/fix|bug|error|issue|broken/.test(text)) preset = 'bugfix'
+
+  return ANGLE_PRESETS[preset].slice(0, count)
+}
+
+const selectedAngles = selectAngles(task_description, complexity === 'High' ? 4 : (complexity === 'Medium' ? 3 : 1))
+
+// Planning strategy determination
+const planningStrategy = complexity === 'Low'
+  ? 'Direct Claude Planning'
+  : 'cli-lite-planning-agent'
+
+console.log(`
+## Exploration Plan
+
+Task Complexity: ${complexity}
+Selected Angles: ${selectedAngles.join(', ')}
+Planning Strategy: ${planningStrategy}
+
+Launching ${selectedAngles.length} parallel explorations...
+`)
+```
+
+**Launch Parallel Explorations** - Orchestrator assigns angle to each agent:
+
+**⚠️ CRITICAL - NO BACKGROUND EXECUTION**:
+- **MUST NOT use `run_in_background: true`** - exploration results are REQUIRED before planning
+
+
+```javascript
+// Launch agents with pre-assigned angles
+const explorationTasks = selectedAngles.map((angle, index) =>
+  Task(
+    subagent_type="cli-explore-agent",
+    run_in_background=false,  // ⚠️ MANDATORY: Must wait for results
+    description=`Explore: ${angle}`,
+    prompt=`
+## Task Objective
+Execute **${angle}** exploration for task planning context. Analyze codebase from this specific angle to discover relevant structure, patterns, and constraints.
+
+## Assigned Context
+- **Exploration Angle**: ${angle}
+- **Task Description**: ${task_description}
+- **Exploration Index**: ${index + 1} of ${selectedAngles.length}
+- **Output File**: ${sessionFolder}/exploration-${angle}.json
+
+## MANDATORY FIRST STEPS (Execute by Agent)
+**You (cli-explore-agent) MUST execute these steps in order:**
+1. Run: ccw tool exec get_modules_by_depth '{}' (project structure)
+2. Run: rg -l "{keyword_from_task}" --type ts (locate relevant files)
+3. Execute: cat ~/.claude/workflows/cli-templates/schemas/explore-json-schema.json (get output schema reference)
+4. Read: .workflow/project-tech.json (technology stack and architecture context)
+5. Read: .workflow/project-guidelines.json (user-defined constraints and conventions)
+
+## Exploration Strategy (${angle} focus)
+
+**Step 1: Structural Scan** (Bash)
+- get_modules_by_depth.sh → identify modules related to ${angle}
+- find/rg → locate files relevant to ${angle} aspect
+- Analyze imports/dependencies from ${angle} perspective
+
+**Step 2: Semantic Analysis** (Gemini CLI)
+- How does existing code handle ${angle} concerns?
+- What patterns are used for ${angle}?
+- Where would new code integrate from ${angle} viewpoint?
+
+**Step 3: Write Output**
+- Consolidate ${angle} findings into JSON
+- Identify ${angle}-specific clarification needs
+
+## Expected Output
+
+**File**: ${sessionFolder}/exploration-${angle}.json
+
+**Schema Reference**: Schema obtained in MANDATORY FIRST STEPS step 3, follow schema exactly
+
+**Required Fields** (all ${angle} focused):
+- project_structure: Modules/architecture relevant to ${angle}
+- relevant_files: Files affected from ${angle} perspective
+  **IMPORTANT**: Use object format with relevance scores for synthesis:
+  \`[{path: "src/file.ts", relevance: 0.85, rationale: "Core ${angle} logic"}]\`
+  Scores: 0.7+ high priority, 0.5-0.7 medium, <0.5 low
+- patterns: ${angle}-related patterns to follow
+- dependencies: Dependencies relevant to ${angle}
+- integration_points: Where to integrate from ${angle} viewpoint (include file:line locations)
+- constraints: ${angle}-specific limitations/conventions
+- clarification_needs: ${angle}-related ambiguities (options array + recommended index)
+- _metadata.exploration_angle: "${angle}"
+
+## Success Criteria
+- [ ] Schema obtained via cat explore-json-schema.json
+- [ ] get_modules_by_depth.sh executed
+- [ ] At least 3 relevant files identified with ${angle} rationale
+- [ ] Patterns are actionable (code examples, not generic advice)
+- [ ] Integration points include file:line locations
+- [ ] Constraints are project-specific to ${angle}
+- [ ] JSON output follows schema exactly
+- [ ] clarification_needs includes options + recommended
+
+## Output
+Write: ${sessionFolder}/exploration-${angle}.json
+Return: 2-3 sentence summary of ${angle} findings
+`
+  )
+)
+
+// Execute all exploration tasks in parallel
+```
+
+**Auto-discover Generated Exploration Files**:
+```javascript
+// After explorations complete, auto-discover all exploration-*.json files
+const explorationFiles = bash(`find ${sessionFolder} -name "exploration-*.json" -type f`)
+  .split('\n')
+  .filter(f => f.trim())
+
+// Read metadata to build manifest
+const explorationManifest = {
+  session_id: sessionId,
+  task_description: task_description,
+  timestamp: getUtc8ISOString(),
+  complexity: complexity,
+  exploration_count: explorationCount,
+  explorations: explorationFiles.map(file => {
+    const data = JSON.parse(Read(file))
+    const filename = path.basename(file)
+    return {
+      angle: data._metadata.exploration_angle,
+      file: filename,
+      path: file,
+      index: data._metadata.exploration_index
+    }
+  })
+}
+
+Write(`${sessionFolder}/explorations-manifest.json`, JSON.stringify(explorationManifest, null, 2))
+
+console.log(`
+## Exploration Complete
+
+Generated exploration files in ${sessionFolder}:
+${explorationManifest.explorations.map(e => `- exploration-${e.angle}.json (angle: ${e.angle})`).join('\n')}
+
+Manifest: explorations-manifest.json
+Angles explored: ${explorationManifest.explorations.map(e => e.angle).join(', ')}
+`)
+```
+
+**Output**:
+- `${sessionFolder}/exploration-{angle1}.json`
+- `${sessionFolder}/exploration-{angle2}.json`
+- ... (1-4 files based on complexity)
+- `${sessionFolder}/explorations-manifest.json`
+
+---
+
+### Phase 2: Clarification (Optional, Multi-Round)
+
+**Skip if**: No exploration or `clarification_needs` is empty across all explorations
+
+**⚠️ CRITICAL**: AskUserQuestion tool limits max 4 questions per call. **MUST execute multiple rounds** to exhaust all clarification needs - do NOT stop at round 1.
+
+**Aggregate clarification needs from all exploration angles**:
+```javascript
+// Load manifest and all exploration files
+const manifest = JSON.parse(Read(`${sessionFolder}/explorations-manifest.json`))
+const explorations = manifest.explorations.map(exp => ({
+  angle: exp.angle,
+  data: JSON.parse(Read(exp.path))
+}))
+
+// Aggregate clarification needs from all explorations
+const allClarifications = []
+explorations.forEach(exp => {
+  if (exp.data.clarification_needs?.length > 0) {
+    exp.data.clarification_needs.forEach(need => {
+      allClarifications.push({
+        ...need,
+        source_angle: exp.angle
+      })
+    })
+  }
+})
+
+// Intelligent deduplication: analyze allClarifications by intent
+// - Identify questions with similar intent across different angles
+// - Merge similar questions: combine options, consolidate context
+// - Produce dedupedClarifications with unique intents only
+const dedupedClarifications = intelligentMerge(allClarifications)
+
+// Multi-round clarification: batch questions (max 4 per round)
+if (dedupedClarifications.length > 0) {
+  const BATCH_SIZE = 4
+  const totalRounds = Math.ceil(dedupedClarifications.length / BATCH_SIZE)
+
+  for (let i = 0; i < dedupedClarifications.length; i += BATCH_SIZE) {
+    const batch = dedupedClarifications.slice(i, i + BATCH_SIZE)
+    const currentRound = Math.floor(i / BATCH_SIZE) + 1
+
+    console.log(`### Clarification Round ${currentRound}/${totalRounds}`)
+
+    AskUserQuestion({
+      questions: batch.map(need => ({
+        question: `[${need.source_angle}] ${need.question}\n\nContext: ${need.context}`,
+        header: need.source_angle.substring(0, 12),
+        multiSelect: false,
+        options: need.options.map((opt, index) => ({
+          label: need.recommended === index ? `${opt} ★` : opt,
+          description: need.recommended === index ? `Recommended` : `Use ${opt}`
+        }))
+      }))
+    })
+
+    // Store batch responses in clarificationContext before next round
+  }
+}
+```
+
+**Output**: `clarificationContext` (in-memory)
+
+---
+
+### Phase 3: Planning
+
+**Planning Strategy Selection** (based on Phase 1 complexity):
+
+**IMPORTANT**: Phase 3 is **planning only** - NO code execution. All execution happens in Phase 5 via lite-execute.
+
+**Executor Assignment** (Claude 智能分配，plan 生成后执行):
+
+```javascript
+// 分配规则（优先级从高到低）：
+// 1. 用户明确指定："用 gemini 分析..." → gemini, "codex 实现..." → codex
+// 2. 默认 → agent
+
+const executorAssignments = {}  // { taskId: { executor: 'gemini'|'codex'|'agent', reason: string } }
+plan.tasks.forEach(task => {
+  // Claude 根据上述规则语义分析，为每个 task 分配 executor
+  executorAssignments[task.id] = { executor: '...', reason: '...' }
+})
+```
+
+**Low Complexity** - Direct planning by Claude:
+```javascript
+// Step 1: Read schema
+const schema = Bash(`cat ~/.claude/workflows/cli-templates/schemas/plan-json-schema.json`)
+
+// Step 2: ⚠️ MANDATORY - Read and review ALL exploration files
+const manifest = JSON.parse(Read(`${sessionFolder}/explorations-manifest.json`))
+manifest.explorations.forEach(exp => {
+  const explorationData = Read(exp.path)
+  console.log(`\n### Exploration: ${exp.angle}\n${explorationData}`)
+})
+
+// Step 3: Generate plan following schema (Claude directly, no agent)
+// ⚠️ Plan MUST incorporate insights from exploration files read in Step 2
+const plan = {
+  summary: "...",
+  approach: "...",
+  tasks: [...],  // Each task: { id, title, scope, ..., depends_on, execution_group, complexity }
+  estimated_time: "...",
+  recommended_execution: "Agent",
+  complexity: "Low",
+  _metadata: { timestamp: getUtc8ISOString(), source: "direct-planning", planning_mode: "direct" }
+}
+
+// Step 4: Write plan to session folder
+Write(`${sessionFolder}/plan.json`, JSON.stringify(plan, null, 2))
+
+// Step 5: MUST continue to Phase 4 (Confirmation) - DO NOT execute code here
+```
+
+**Medium/High Complexity** - Invoke cli-lite-planning-agent:
+
+```javascript
+Task(
+  subagent_type="cli-lite-planning-agent",
+  run_in_background=false,
+  description="Generate detailed implementation plan",
+  prompt=`
+Generate implementation plan and write plan.json.
+
+## Output Schema Reference
+Execute: cat ~/.claude/workflows/cli-templates/schemas/plan-json-schema.json (get schema reference before generating plan)
+
+## Project Context (MANDATORY - Read Both Files)
+1. Read: .workflow/project-tech.json (technology stack, architecture, key components)
+2. Read: .workflow/project-guidelines.json (user-defined constraints and conventions)
+
+**CRITICAL**: All generated tasks MUST comply with constraints in project-guidelines.json
+
+## Task Description
+${task_description}
+
+## Multi-Angle Exploration Context
+
+${manifest.explorations.map(exp => `### Exploration: ${exp.angle} (${exp.file})
+Path: ${exp.path}
+
+Read this file for detailed ${exp.angle} analysis.`).join('\n\n')}
+
+Total explorations: ${manifest.exploration_count}
+Angles covered: ${manifest.explorations.map(e => e.angle).join(', ')}
+
+Manifest: ${sessionFolder}/explorations-manifest.json
+
+## User Clarifications
+${JSON.stringify(clarificationContext) || "None"}
+
+## Complexity Level
+${complexity}
+
+## Requirements
+Generate plan.json following the schema obtained above. Key constraints:
+- tasks: 2-7 structured tasks (**group by feature/module, NOT by file**)
+- _metadata.exploration_angles: ${JSON.stringify(manifest.explorations.map(e => e.angle))}
+
+## Task Grouping Rules
+1. **Group by feature**: All changes for one feature = one task (even if 3-5 files)
+2. **Group by context**: Tasks with similar context or related functional changes can be grouped together
+3. **Minimize agent count**: Simple, unrelated tasks can also be grouped to reduce agent execution overhead
+4. **Avoid file-per-task**: Do NOT create separate tasks for each file
+5. **Substantial tasks**: Each task should represent 15-60 minutes of work
+6. **True dependencies only**: Only use depends_on when Task B cannot start without Task A's output
+7. **Prefer parallel**: Most tasks should be independent (no depends_on)
+
+## Execution
+1. Read schema file (cat command above)
+2. Execute CLI planning using Gemini (Qwen fallback)
+3. Read ALL exploration files for comprehensive context
+4. Synthesize findings and generate plan following schema
+5. Write JSON: Write('${sessionFolder}/plan.json', jsonContent)
+6. Return brief completion summary
+`
+)
+```
+
+**Output**: `${sessionFolder}/plan.json`
+
+---
+
+### Phase 4: Task Confirmation & Execution Selection
+
+**Step 4.1: Display Plan**
+```javascript
+const plan = JSON.parse(Read(`${sessionFolder}/plan.json`))
+
+console.log(`
+## Implementation Plan
+
+**Summary**: ${plan.summary}
+**Approach**: ${plan.approach}
+
+**Tasks** (${plan.tasks.length}):
+${plan.tasks.map((t, i) => `${i+1}. ${t.title} (${t.file})`).join('\n')}
+
+**Complexity**: ${plan.complexity}
+**Estimated Time**: ${plan.estimated_time}
+**Recommended**: ${plan.recommended_execution}
+`)
+```
+
+**Step 4.2: Collect Confirmation**
+```javascript
+AskUserQuestion({
+  questions: [
+    {
+      question: `Confirm plan? (${plan.tasks.length} tasks, ${plan.complexity})`,
+      header: "Confirm",
+      multiSelect: true,
+      options: [
+        { label: "Allow", description: "Proceed as-is" },
+        { label: "Modify", description: "Adjust before execution" },
+        { label: "Cancel", description: "Abort workflow" }
+      ]
+    },
+    {
+      question: "Execution method:",
+      header: "Execution",
+      multiSelect: false,
+      options: [
+        { label: "Agent", description: "@code-developer agent" },
+        { label: "Codex", description: "codex CLI tool" },
+        { label: "Auto", description: `Auto: ${plan.complexity === 'Low' ? 'Agent' : 'Codex'}` }
+      ]
+    },
+    {
+      question: "Code review after execution?",
+      header: "Review",
+      multiSelect: false,
+      options: [
+        { label: "Gemini Review", description: "Gemini CLI" },
+        { label: "Agent Review", description: "@code-reviewer" },
+        { label: "Skip", description: "No review" }
+      ]
+    }
+  ]
+})
+```
+
+---
+
+### Phase 5: Dispatch to Execution
+
+**CRITICAL**: lite-plan NEVER executes code directly. ALL execution MUST go through lite-execute.
+
+**Step 5.1: Build executionContext**
+
+```javascript
+// Load manifest and all exploration files
+const manifest = JSON.parse(Read(`${sessionFolder}/explorations-manifest.json`))
+const explorations = {}
+
+manifest.explorations.forEach(exp => {
+  if (file_exists(exp.path)) {
+    explorations[exp.angle] = JSON.parse(Read(exp.path))
+  }
+})
+
+const plan = JSON.parse(Read(`${sessionFolder}/plan.json`))
+
+executionContext = {
+  planObject: plan,
+  explorationsContext: explorations,
+  explorationAngles: manifest.explorations.map(e => e.angle),
+  explorationManifest: manifest,
+  clarificationContext: clarificationContext || null,
+  executionMethod: userSelection.execution_method,  // 全局默认，可被 executorAssignments 覆盖
+  codeReviewTool: userSelection.code_review_tool,
+  originalUserInput: task_description,
+
+  // 任务级 executor 分配（优先于全局 executionMethod）
+  executorAssignments: executorAssignments,  // { taskId: { executor, reason } }
+
+  session: {
+    id: sessionId,
+    folder: sessionFolder,
+    artifacts: {
+      explorations: manifest.explorations.map(exp => ({
+        angle: exp.angle,
+        path: exp.path
+      })),
+      explorations_manifest: `${sessionFolder}/explorations-manifest.json`,
+      plan: `${sessionFolder}/plan.json`
+    }
+  }
+}
+```
+
+**Step 5.2: Dispatch**
+
+```javascript
+SlashCommand(command="/workflow:lite-execute --in-memory")
+```
+
+## Session Folder Structure
+
+```
+.workflow/.lite-plan/{task-slug}-{YYYY-MM-DD}/
+├── exploration-{angle1}.json      # Exploration angle 1
+├── exploration-{angle2}.json      # Exploration angle 2
+├── exploration-{angle3}.json      # Exploration angle 3 (if applicable)
+├── exploration-{angle4}.json      # Exploration angle 4 (if applicable)
+├── explorations-manifest.json     # Exploration index
+└── plan.json                      # Implementation plan
+```
+
+**Example**:
+```
+.workflow/.lite-plan/implement-jwt-refresh-2025-11-25-14-30-25/
+├── exploration-architecture.json
+├── exploration-auth-patterns.json
+├── exploration-security.json
+├── explorations-manifest.json
+└── plan.json
+```
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| Exploration agent failure | Skip exploration, continue with task description only |
+| Planning agent failure | Fallback to direct planning by Claude |
+| Clarification timeout | Use exploration findings as-is |
+| Confirmation timeout | Save context, display resume instructions |
+| Modify loop > 3 times | Suggest breaking task or using /workflow:plan |

.claude/commands/workflow/plan.md

@@ -1,7 +1,7 @@
 ---
 name: plan
-description: Orchestrate 5-phase planning workflow with quality gate, executing commands and passing context between phases
-argument-hint: "[--agent] [--cli-execute] \"text description\"|file.md"
+description: 5-phase planning workflow with action-planning-agent task generation, outputs IMPL_PLAN.md and task JSONs
+argument-hint: "\"text description\"|file.md"
 allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 ---
 
@@ -9,7 +9,7 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 
 ## Coordinator Role
 
-**This command is a pure orchestrator**: Execute 5 slash commands in sequence (including a quality gate), parse their outputs, pass context between them, and ensure complete execution through **automatic continuation**.
+**This command is a pure orchestrator**: Dispatch 5 slash commands in sequence (including a quality gate), parse their outputs, pass context between them, and ensure complete execution through **automatic continuation**.
 
 **Execution Model - Auto-Continue Workflow with Quality Gate**:
 
@@ -17,21 +17,24 @@ This workflow runs **fully autonomously** once triggered. Phase 3 (conflict reso
 
 
 1. **User triggers**: `/workflow:plan "task"`
-2. **Phase 1 executes** → Session discovery → Auto-continues
-3. **Phase 2 executes** → Context gathering → Auto-continues
-4. **Phase 3 executes** (optional, if conflict_risk ≥ medium) → Conflict resolution → Auto-continues
-5. **Phase 4 executes** (task-generate-agent if --agent) → Task generation → Reports final summary
+2. **Phase 1 dispatches** → Session discovery → Auto-continues
+3. **Phase 2 dispatches** → Context gathering → Auto-continues
+4. **Phase 3 dispatches** (optional, if conflict_risk ≥ medium) → Conflict resolution → Auto-continues
+5. **Phase 4 dispatches** → Task generation (task-generate-agent) → Reports final summary
+
+**Task Attachment Model**:
+- SlashCommand dispatch **expands workflow** by attaching sub-tasks to current TodoWrite
+- When a sub-command is dispatched (e.g., `/workflow:tools:context-gather`), its internal tasks are attached to the orchestrator's TodoWrite
+- Orchestrator **executes these attached tasks** sequentially
+- After completion, attached tasks are **collapsed** back to high-level phase summary
+- This is **task expansion**, not external delegation
 
 **Auto-Continue Mechanism**:
-- TodoList tracks current phase status
-- After each phase completion, automatically executes next pending phase
+- TodoList tracks current phase status and dynamically manages task attachment/collapse
+- When each phase finishes executing, automatically execute next pending phase
 - All phases run autonomously without user interaction (clarification handled in brainstorm phase)
 - Progress updates shown at each phase for visibility
-
-**Execution Modes**:
-- **Manual Mode** (default): Use `/workflow:tools:task-generate`
-- **Agent Mode** (`--agent`): Use `/workflow:tools:task-generate-agent`
-- **CLI Execute Mode** (`--cli-execute`): Generate tasks with Codex execution commands
+- **⚠️ CONTINUOUS EXECUTION** - Do not stop until all phases complete
 
 ## Core Rules
 
@@ -39,13 +42,49 @@ This workflow runs **fully autonomously** once triggered. Phase 3 (conflict reso
 2. **No Preliminary Analysis**: Do not read files, analyze structure, or gather context before Phase 1
 3. **Parse Every Output**: Extract required data from each command/agent output for next phase
 4. **Auto-Continue via TodoList**: Check TodoList status to execute next pending phase automatically
-5. **Track Progress**: Update TodoWrite after every phase completion
-6. **Agent Delegation**: Phase 3 uses cli-execution-agent for autonomous intelligent analysis
+5. **Track Progress**: Update TodoWrite dynamically with task attachment/collapse pattern
+6. **Task Attachment Model**: SlashCommand dispatch **attaches** sub-tasks to current workflow. Orchestrator **executes** these attached tasks itself, then **collapses** them after completion
+7. **⚠️ CRITICAL: DO NOT STOP**: Continuous multi-phase workflow. After executing all attached tasks, immediately collapse them and execute next phase
+
+## Execution Process
+
+```
+Input Parsing:
+   └─ Convert user input to structured format (GOAL/SCOPE/CONTEXT)
+
+Phase 1: Session Discovery
+   └─ /workflow:session:start --auto "structured-description"
+      └─ Output: sessionId (WFS-xxx)
+
+Phase 2: Context Gathering
+   └─ /workflow:tools:context-gather --session sessionId "structured-description"
+      ├─ Tasks attached: Analyze structure → Identify integration → Generate package
+      └─ Output: contextPath + conflict_risk
+
+Phase 3: Conflict Resolution 
+   └─ Decision (conflict_risk check):
+      ├─ conflict_risk ≥ medium → Execute /workflow:tools:conflict-resolution
+      │   ├─ Tasks attached: Detect conflicts → Present to user → Apply strategies
+      │   └─ Output: Modified brainstorm artifacts
+      └─ conflict_risk < medium → Skip to Phase 4
+
+Phase 4: Task Generation
+   └─ /workflow:tools:task-generate-agent --session sessionId
+      └─ Output: IMPL_PLAN.md, task JSONs, TODO_LIST.md
+
+Return:
+   └─ Summary with recommended next steps
+```
 
 ## 5-Phase Execution
 
 ### Phase 1: Session Discovery
-**Command**: `SlashCommand(command="/workflow:session:start --auto \"[structured-task-description]\"")`
+
+**Step 1.1: Dispatch** - Create or discover workflow session
+
+```javascript
+SlashCommand(command="/workflow:session:start --auto \"[structured-task-description]\"")
+```
 
 **Task Description Structure**:
 ```
@@ -66,7 +105,9 @@ CONTEXT: Existing user database schema, REST API endpoints
 
 **Validation**:
 - Session ID successfully extracted
-- Session directory `.workflow/[sessionId]/` exists
+- Session directory `.workflow/active/[sessionId]/` exists
+
+**Note**: Session directory contains `workflow-session.json` (metadata). Do NOT look for `manifest.json` here - it only exists in `.workflow/archives/` for archived sessions.
 
 **TodoWrite**: Mark phase 1 completed, phase 2 in_progress
 
@@ -75,7 +116,12 @@ CONTEXT: Existing user database schema, REST API endpoints
 ---
 
 ### Phase 2: Context Gathering
-**Command**: `SlashCommand(command="/workflow:tools:context-gather --session [sessionId] \"[structured-task-description]\"")`
+
+**Step 2.1: Dispatch** - Gather project context and analyze codebase
+
+```javascript
+SlashCommand(command="/workflow:tools:context-gather --session [sessionId] \"[structured-task-description]\"")
+```
 
 **Use Same Structured Description**: Pass the same structured format from Phase 1
 
@@ -83,23 +129,54 @@ CONTEXT: Existing user database schema, REST API endpoints
 
 **Parse Output**:
 - Extract: context-package.json path (store as `contextPath`)
-- Typical pattern: `.workflow/[sessionId]/.process/context-package.json`
+- Typical pattern: `.workflow/active/[sessionId]/.process/context-package.json`
 
 **Validation**:
 - Context package path extracted
 - File exists and is valid JSON
 
-**TodoWrite**: Mark phase 2 completed, phase 3 in_progress
+<!-- TodoWrite: When context-gather dispatched, INSERT 3 context-gather tasks, mark first as in_progress -->
 
-**After Phase 2**: Return to user showing Phase 2 results, then auto-continue to Phase 3
+**TodoWrite Update (Phase 2 SlashCommand dispatched - tasks attached)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "in_progress", "activeForm": "Executing context gathering"},
+  {"content": "  → Analyze codebase structure", "status": "in_progress", "activeForm": "Analyzing codebase structure"},
+  {"content": "  → Identify integration points", "status": "pending", "activeForm": "Identifying integration points"},
+  {"content": "  → Generate context package", "status": "pending", "activeForm": "Generating context package"},
+  {"content": "Phase 4: Task Generation", "status": "pending", "activeForm": "Executing task generation"}
+]
+```
+
+**Note**: SlashCommand dispatch **attaches** context-gather's 3 tasks. Orchestrator **executes** these tasks sequentially.
+
+<!-- TodoWrite: After Phase 2 tasks complete, REMOVE Phase 2.1-2.3, restore to orchestrator view -->
+
+**TodoWrite Update (Phase 2 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 4: Task Generation", "status": "pending", "activeForm": "Executing task generation"}
+]
+```
+
+**Note**: Phase 2 tasks completed and collapsed to summary.
+
+**After Phase 2**: Return to user showing Phase 2 results, then auto-continue to Phase 3/4 (depending on conflict_risk)
 
 ---
 
-### Phase 3: Conflict Resolution (Optional - auto-triggered by conflict risk)
+### Phase 3: Conflict Resolution
 
 **Trigger**: Only execute when context-package.json indicates conflict_risk is "medium" or "high"
 
-**Command**: `SlashCommand(command="/workflow:tools:conflict-resolution --session [sessionId] --context [contextPath]")`
+**Step 3.1: Dispatch** - Detect and resolve conflicts with CLI analysis
+
+```javascript
+SlashCommand(command="/workflow:tools:conflict-resolution --session [sessionId] --context [contextPath]")
+```
 
 **Input**:
 - sessionId from Phase 1
@@ -108,24 +185,58 @@ CONTEXT: Existing user database schema, REST API endpoints
 
 **Parse Output**:
 - Extract: Execution status (success/skipped/failed)
-- Verify: CONFLICT_RESOLUTION.md file path (if executed)
+- Verify: conflict-resolution.json file path (if executed)
 
 **Validation**:
-- File `.workflow/[sessionId]/.process/CONFLICT_RESOLUTION.md` exists (if executed)
+- File `.workflow/active/[sessionId]/.process/conflict-resolution.json` exists (if executed)
 
 **Skip Behavior**:
 - If conflict_risk is "none" or "low", skip directly to Phase 3.5
 - Display: "No significant conflicts detected, proceeding to clarification"
 
-**TodoWrite**: Mark phase 3 completed (if executed) or skipped, phase 3.5 in_progress
+<!-- TodoWrite: If conflict_risk ≥ medium, INSERT 3 conflict-resolution tasks -->
+
+**TodoWrite Update (Phase 3 SlashCommand dispatched - tasks attached, if conflict_risk ≥ medium)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 3: Conflict Resolution", "status": "in_progress", "activeForm": "Resolving conflicts"},
+  {"content": "  → Detect conflicts with CLI analysis", "status": "in_progress", "activeForm": "Detecting conflicts"},
+  {"content": "  → Present conflicts to user", "status": "pending", "activeForm": "Presenting conflicts"},
+  {"content": "  → Apply resolution strategies", "status": "pending", "activeForm": "Applying resolution strategies"},
+  {"content": "Phase 4: Task Generation", "status": "pending", "activeForm": "Executing task generation"}
+]
+```
+
+**Note**: SlashCommand dispatch **attaches** conflict-resolution's 3 tasks. Orchestrator **executes** these tasks sequentially.
+
+<!-- TodoWrite: After Phase 3 tasks complete, REMOVE Phase 3.1-3.3, restore to orchestrator view -->
+
+**TodoWrite Update (Phase 3 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 3: Conflict Resolution", "status": "completed", "activeForm": "Resolving conflicts"},
+  {"content": "Phase 4: Task Generation", "status": "pending", "activeForm": "Executing task generation"}
+]
+```
+
+**Note**: Phase 3 tasks completed and collapsed to summary.
 
 **After Phase 3**: Return to user showing conflict resolution results (if executed) and selected strategies, then auto-continue to Phase 3.5
 
 **Memory State Check**:
 - Evaluate current context window usage and memory state
-- If memory usage is high (>110K tokens or approaching context limits):
-  - **Command**: `SlashCommand(command="/compact")`
-  - This optimizes memory before proceeding to Phase 3.5
+- If memory usage is high (>120K tokens or approaching context limits):
+
+  **Step 3.2: Dispatch** - Optimize memory before proceeding
+
+  ```javascript
+  SlashCommand(command="/compact")
+  ```
+
 - Memory compaction is particularly important after analysis phase which may generate extensive documentation
 - Ensures optimal performance and prevents context overflow
 
@@ -159,38 +270,52 @@ CONTEXT: Existing user database schema, REST API endpoints
 - Task generation translates high-level role analyses into concrete, actionable work items
 - **Intent priority**: Current user prompt > role analysis.md files > guidance-specification.md
 
-**Command Selection**:
-- Manual: `SlashCommand(command="/workflow:tools:task-generate --session [sessionId]")`
-- Agent: `SlashCommand(command="/workflow:tools:task-generate-agent --session [sessionId]")`
-- CLI Execute: Add `--cli-execute` flag to either command
+**Step 4.1: Dispatch** - Generate implementation plan and task JSONs
 
-**Flag Combination**:
-- `--cli-execute` alone: Manual task generation with CLI execution
-- `--agent --cli-execute`: Agent task generation with CLI execution
-
-**Command Examples**:
-```bash
-# Manual with CLI execution
-/workflow:tools:task-generate --session WFS-auth --cli-execute
-
-# Agent with CLI execution
-/workflow:tools:task-generate-agent --session WFS-auth --cli-execute
+```javascript
+SlashCommand(command="/workflow:tools:task-generate-agent --session [sessionId]")
 ```
 
+**CLI Execution Note**: CLI tool usage is now determined semantically by action-planning-agent based on user's task description. If user specifies "use Codex/Gemini/Qwen for X", the agent embeds `command` fields in relevant `implementation_approach` steps.
+
 **Input**: `sessionId` from Phase 1
 
 **Validation**:
-- `.workflow/[sessionId]/IMPL_PLAN.md` exists
-- `.workflow/[sessionId]/.task/IMPL-*.json` exists (at least one)
-- `.workflow/[sessionId]/TODO_LIST.md` exists
+- `.workflow/active/[sessionId]/IMPL_PLAN.md` exists
+- `.workflow/active/[sessionId]/.task/IMPL-*.json` exists (at least one)
+- `.workflow/active/[sessionId]/TODO_LIST.md` exists
 
-**TodoWrite**: Mark phase 4 completed
+<!-- TodoWrite: When task-generate-agent dispatched, ATTACH 1 agent task -->
+
+**TodoWrite Update (Phase 4 SlashCommand dispatched - agent task attached)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 4: Task Generation", "status": "in_progress", "activeForm": "Executing task generation"}
+]
+```
+
+**Note**: Single agent task attached. Agent autonomously completes discovery, planning, and output generation internally.
+
+<!-- TodoWrite: After agent completes, mark task as completed -->
+
+**TodoWrite Update (Phase 4 completed)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 4: Task Generation", "status": "completed", "activeForm": "Executing task generation"}
+]
+```
+
+**Note**: Agent task completed. No collapse needed (single task).
 
 **Return to User**:
 ```
 Planning complete for session: [sessionId]
 Tasks generated: [count]
-Plan: .workflow/[sessionId]/IMPL_PLAN.md
+Plan: .workflow/active/[sessionId]/IMPL_PLAN.md
 
 Recommended Next Steps:
 1. /workflow:action-plan-verify --session [sessionId]  # Verify plan quality before execution
@@ -202,39 +327,36 @@ Quality Gate: Consider running /workflow:action-plan-verify to catch issues earl
 
 ## TodoWrite Pattern
 
-```javascript
-// Initialize (before Phase 1)
-// Note: Phase 3 todo only added dynamically after Phase 2 if conflict_risk ≥ medium
-TodoWrite({todos: [
-  {"content": "Execute session discovery", "status": "in_progress", "activeForm": "Executing session discovery"},
-  {"content": "Execute context gathering", "status": "pending", "activeForm": "Executing context gathering"},
-  // Phase 3 todo added dynamically after Phase 2 if conflict_risk ≥ medium
-  {"content": "Execute task generation", "status": "pending", "activeForm": "Executing task generation"}
-]})
+**Core Concept**: Dynamic task attachment and collapse for real-time visibility into workflow execution.
 
-// After Phase 2 (if conflict_risk ≥ medium, insert Phase 3 todo)
-TodoWrite({todos: [
-  {"content": "Execute session discovery", "status": "completed", "activeForm": "Executing session discovery"},
-  {"content": "Execute context gathering", "status": "completed", "activeForm": "Executing context gathering"},
-  {"content": "Resolve conflicts and apply fixes", "status": "in_progress", "activeForm": "Resolving conflicts"},
-  {"content": "Execute task generation", "status": "pending", "activeForm": "Executing task generation"}
-]})
+### Key Principles
 
-// After Phase 2 (if conflict_risk is none/low, skip Phase 3, go directly to Phase 4)
-TodoWrite({todos: [
-  {"content": "Execute session discovery", "status": "completed", "activeForm": "Executing session discovery"},
-  {"content": "Execute context gathering", "status": "completed", "activeForm": "Executing context gathering"},
-  {"content": "Execute task generation", "status": "in_progress", "activeForm": "Executing task generation"}
-]})
+1. **Task Attachment** (when SlashCommand dispatched):
+   - Sub-command's internal tasks are **attached** to orchestrator's TodoWrite
+   - **Phase 2, 3**: Multiple sub-tasks attached (e.g., Phase 2.1, 2.2, 2.3)
+   - **Phase 4**: Single agent task attached (e.g., "Execute task-generate-agent")
+   - First attached task marked as `in_progress`, others as `pending`
+   - Orchestrator **executes** these attached tasks sequentially
 
-// After Phase 3 (if executed), continue to Phase 4
-TodoWrite({todos: [
-  {"content": "Execute session discovery", "status": "completed", "activeForm": "Executing session discovery"},
-  {"content": "Execute context gathering", "status": "completed", "activeForm": "Executing context gathering"},
-  {"content": "Resolve conflicts and apply fixes", "status": "completed", "activeForm": "Resolving conflicts"},
-  {"content": "Execute task generation", "status": "in_progress", "activeForm": "Executing task generation"}
-]})
-```
+2. **Task Collapse** (after sub-tasks complete):
+   - **Applies to Phase 2, 3**: Remove detailed sub-tasks from TodoWrite
+   - **Collapse** to high-level phase summary
+   - Example: Phase 2.1-2.3 collapse to "Execute context gathering: completed"
+   - **Phase 4**: No collapse needed (single task, just mark completed)
+   - Maintains clean orchestrator-level view
+
+3. **Continuous Execution**:
+   - After completion, automatically proceed to next pending phase
+   - No user intervention required between phases
+   - TodoWrite dynamically reflects current execution state
+
+**Lifecycle Summary**: Initial pending tasks → Phase dispatched (tasks ATTACHED) → Sub-tasks executed sequentially → Phase completed (tasks COLLAPSED to summary for Phase 2/3, or marked completed for Phase 4) → Next phase begins → Repeat until all phases complete.
+
+
+
+**Note**: See individual Phase descriptions for detailed TodoWrite Update examples:
+- **Phase 2, 3**: Multiple sub-tasks with attach/collapse pattern
+- **Phase 4**: Single agent task (no collapse needed)
 
 ## Input Processing
 
@@ -296,7 +418,7 @@ Phase 3: conflict-resolution [AUTO-TRIGGERED if conflict_risk ≥ medium]
     ↓ Output: Modified brainstorm artifacts (NO report file)
     ↓ Skip if conflict_risk is none/low → proceed directly to Phase 4
     ↓
-Phase 4: task-generate[--agent] --session sessionId
+Phase 4: task-generate-agent --session sessionId
     ↓ Input: sessionId + resolved brainstorm artifacts + session memory
     ↓ Output: IMPL_PLAN.md, task JSONs, TODO_LIST.md
     ↓
@@ -309,11 +431,55 @@ Return summary to user
 - Brainstorming artifacts (potentially modified by Phase 3)
 - Session-specific configuration
 
-**Structured Description Benefits**:
-- **Clarity**: Clear separation of goal, scope, and context
-- **Consistency**: Same format across all phases
-- **Traceability**: Easy to track what was requested
-- **Precision**: Better context gathering and analysis
+
+## Execution Flow Diagram
+
+```
+User triggers: /workflow:plan "Build authentication system"
+  ↓
+[TodoWrite Init] 3 orchestrator-level tasks
+  ↓
+Phase 1: Session Discovery
+  → sessionId extracted
+  ↓
+Phase 2: Context Gathering (SlashCommand dispatched)
+  → ATTACH 3 sub-tasks: ← ATTACHED
+    - → Analyze codebase structure
+    - → Identify integration points
+    - → Generate context package
+  → Execute sub-tasks sequentially
+  → COLLAPSE tasks ← COLLAPSED
+  → contextPath + conflict_risk extracted
+  ↓
+Conditional Branch: Check conflict_risk
+  ├─ IF conflict_risk ≥ medium:
+  │   Phase 3: Conflict Resolution (SlashCommand dispatched)
+  │     → ATTACH 3 sub-tasks: ← ATTACHED
+  │       - → Detect conflicts with CLI analysis
+  │       - → Present conflicts to user
+  │       - → Apply resolution strategies
+  │     → Execute sub-tasks sequentially
+  │     → COLLAPSE tasks ← COLLAPSED
+  │
+  └─ ELSE: Skip Phase 3, proceed to Phase 4
+  ↓
+Phase 4: Task Generation (SlashCommand dispatched)
+  → Single agent task (no sub-tasks)
+  → Agent autonomously completes internally:
+    (discovery → planning → output)
+  → Outputs: IMPL_PLAN.md, IMPL-*.json, TODO_LIST.md
+  ↓
+Return summary to user
+```
+
+**Key Points**:
+- **← ATTACHED**: Tasks attached to TodoWrite when SlashCommand dispatched
+  - Phase 2, 3: Multiple sub-tasks
+  - Phase 4: Single agent task
+- **← COLLAPSED**: Sub-tasks collapsed to summary after completion (Phase 2, 3 only)
+- **Phase 4**: Single agent task, no collapse (just mark completed)
+- **Conditional Branch**: Phase 3 only dispatches if conflict_risk ≥ medium
+- **Continuous Flow**: No user intervention between phases
 
 ## Error Handling
 
@@ -331,12 +497,9 @@ Return summary to user
 - Parse context path from Phase 2 output, store in memory
 - **Extract conflict_risk from context-package.json**: Determine Phase 3 execution
 - **If conflict_risk ≥ medium**: Launch Phase 3 conflict-resolution with sessionId and contextPath
-- Wait for Phase 3 completion (if executed), verify CONFLICT_RESOLUTION.md created
+- Wait for Phase 3 to finish executing (if executed), verify conflict-resolution.json created
 - **If conflict_risk is none/low**: Skip Phase 3, proceed directly to Phase 4
-- **Build Phase 4 command** based on flags:
-  - Base command: `/workflow:tools:task-generate` (or `-agent` if `--agent` flag)
-  - Add `--session [sessionId]`
-  - Add `--cli-execute` if flag present
+- **Build Phase 4 command**: `/workflow:tools:task-generate-agent --session [sessionId]`
 - Pass session ID to Phase 4 command
 - Verify all Phase 4 outputs
 - Update TodoWrite after each phase (dynamically adjust for Phase 3 presence)
@@ -380,8 +543,7 @@ CONSTRAINTS: [Limitations or boundaries]
 - `/workflow:tools:context-gather` - Phase 2: Gather project context and analyze codebase
 - `/workflow:tools:conflict-resolution` - Phase 3: Detect and resolve conflicts (auto-triggered if conflict_risk ≥ medium)
 - `/compact` - Phase 3: Memory optimization (if context approaching limits)
-- `/workflow:tools:task-generate` - Phase 4: Generate task JSON files with manual approach
-- `/workflow:tools:task-generate-agent` - Phase 4: Generate task JSON files with agent-driven approach (when `--agent` flag used)
+- `/workflow:tools:task-generate-agent` - Phase 4: Generate task JSON files with agent-driven approach
 
 **Follow-up Commands**:
 - `/workflow:action-plan-verify` - Recommended: Verify plan quality and catch issues before execution

.claude/commands/workflow/replan.md

@@ -0,0 +1,515 @@
+---
+name: replan
+description: Interactive workflow replanning with session-level artifact updates and boundary clarification through guided questioning
+argument-hint: "[--session session-id] [task-id] \"requirements\"|file.md [--interactive]"
+allowed-tools: Read(*), Write(*), Edit(*), TodoWrite(*), Glob(*), Bash(*)
+---
+
+# Workflow Replan Command
+
+## Overview
+Intelligently replans workflow sessions or individual tasks with interactive boundary clarification and comprehensive artifact updates.
+
+**Core Capabilities**:
+- **Session Replan**: Updates multiple artifacts (IMPL_PLAN.md, TODO_LIST.md, task JSONs)
+- **Task Replan**: Focused updates within session context
+- **Interactive Clarification**: Guided questioning to define modification boundaries
+- **Impact Analysis**: Automatic detection of affected files and dependencies
+- **Backup Management**: Preserves previous versions with restore capability
+
+## Operation Modes
+
+### Session Replan Mode
+
+```bash
+# Auto-detect active session
+/workflow:replan "添加双因素认证支持"
+
+# Explicit session
+/workflow:replan --session WFS-oauth "添加双因素认证支持"
+
+# File-based input
+/workflow:replan --session WFS-oauth requirements-update.md
+
+# Interactive mode
+/workflow:replan --interactive
+```
+
+### Task Replan Mode
+
+```bash
+# Direct task update
+/workflow:replan IMPL-1 "修改为使用 OAuth2.0 标准"
+
+# Task with explicit session
+/workflow:replan --session WFS-oauth IMPL-2 "增加单元测试覆盖率到 90%"
+
+# Interactive mode
+/workflow:replan IMPL-1 --interactive
+```
+
+## Execution Process
+
+```
+Input Parsing:
+   ├─ Parse flags: --session, --interactive
+   └─ Detect mode: task-id present → Task mode | Otherwise → Session mode
+
+Phase 1: Mode Detection & Session Discovery
+   ├─ Detect operation mode (Task vs Session)
+   ├─ Discover/validate session (--session flag or auto-detect)
+   └─ Load session context (workflow-session.json, IMPL_PLAN.md, TODO_LIST.md)
+
+Phase 2: Interactive Requirement Clarification
+   └─ Decision (by mode):
+      ├─ Session mode → 3-4 questions (scope, modules, changes, dependencies)
+      └─ Task mode → 2 questions (update type, ripple effect)
+
+Phase 3: Impact Analysis & Planning
+   ├─ Analyze required changes
+   ├─ Generate modification plan
+   └─ User confirmation (Execute / Adjust / Cancel)
+
+Phase 4: Backup Creation
+   └─ Backup all affected files with manifest
+
+Phase 5: Apply Modifications
+   ├─ Update IMPL_PLAN.md (if needed)
+   ├─ Update TODO_LIST.md (if needed)
+   ├─ Update/Create/Delete task JSONs
+   └─ Update session metadata
+
+Phase 6: Verification & Summary
+   ├─ Validate consistency (JSON validity, task limits, acyclic dependencies)
+   └─ Generate change summary
+```
+
+## Execution Lifecycle
+
+### Input Parsing
+
+**Parse flags**:
+```javascript
+const sessionFlag = $ARGUMENTS.match(/--session\s+(\S+)/)?.[1]
+const interactive = $ARGUMENTS.includes('--interactive')
+const taskIdMatch = $ARGUMENTS.match(/\b(IMPL-\d+(?:\.\d+)?)\b/)
+const taskId = taskIdMatch?.[1]
+```
+
+### Phase 1: Mode Detection & Session Discovery
+
+**Process**:
+1. **Detect Operation Mode**:
+   - Check if task ID provided (IMPL-N or IMPL-N.M format) → Task mode
+   - Otherwise → Session mode
+
+2. **Discover/Validate Session**:
+   - Use `--session` flag if provided
+   - Otherwise auto-detect from `.workflow/active/`
+   - Validate session exists
+
+3. **Load Session Context**:
+   - Read `workflow-session.json`
+   - List existing tasks
+   - Read `IMPL_PLAN.md` and `TODO_LIST.md`
+
+**Output**: Session validated, context loaded, mode determined
+
+---
+
+### Phase 2: Interactive Requirement Clarification
+
+**Purpose**: Define modification scope through guided questioning
+
+#### Session Mode Questions
+
+**Q1: Modification Scope**
+```javascript
+Options:
+- 仅更新任务细节 (tasks_only)
+- 修改规划方案 (plan_update)
+- 重构任务结构 (task_restructure)
+- 全面重规划 (comprehensive)
+```
+
+**Q2: Affected Modules** (if scope >= plan_update)
+```javascript
+Options: Dynamically generated from existing tasks' focus_paths
+- 认证模块 (src/auth)
+- 用户管理 (src/user)
+- 全部模块
+```
+
+**Q3: Task Changes** (if scope >= task_restructure)
+```javascript
+Options:
+- 添加/删除任务 (add_remove)
+- 合并/拆分任务 (merge_split)
+- 仅更新内容 (update_only)
+// Note: Max 4 options for AskUserQuestion
+```
+
+**Q4: Dependency Changes**
+```javascript
+Options:
+- 是,需要重新梳理依赖
+- 否,保持现有依赖
+```
+
+#### Task Mode Questions
+
+**Q1: Update Type**
+```javascript
+Options:
+- 需求和验收标准 (requirements & acceptance)
+- 实现方案 (implementation_approach)
+- 文件范围 (focus_paths)
+- 依赖关系 (depends_on)
+- 全部更新
+```
+
+**Q2: Ripple Effect**
+```javascript
+Options:
+- 是,需要同步更新依赖任务
+- 否,仅影响当前任务
+- 不确定,请帮我分析
+```
+
+**Output**: User selections stored, modification boundaries defined
+
+---
+
+### Phase 3: Impact Analysis & Planning
+
+**Step 3.1: Analyze Required Changes**
+
+Determine affected files based on clarification:
+
+```typescript
+interface ImpactAnalysis {
+  affected_files: {
+    impl_plan: boolean;
+    todo_list: boolean;
+    session_meta: boolean;
+    tasks: string[];
+  };
+
+  operations: {
+    type: 'create' | 'update' | 'delete' | 'merge' | 'split';
+    target: string;
+    reason: string;
+  }[];
+
+  backup_strategy: {
+    timestamp: string;
+    files: string[];
+  };
+}
+```
+
+**Step 3.2: Generate Modification Plan**
+
+```markdown
+## 修改计划
+
+### 影响范围
+- [ ] IMPL_PLAN.md: 更新技术方案第 3 节
+- [ ] TODO_LIST.md: 添加 2 个新任务,删除 1 个废弃任务
+- [ ] IMPL-001.json: 更新实现方案
+- [ ] workflow-session.json: 更新任务计数
+
+### 变更操作
+1. **创建**: IMPL-004.json (双因素认证实现)
+2. **更新**: IMPL-001.json (添加 2FA 准备工作)
+3. **删除**: IMPL-003.json (已被新方案替代)
+```
+
+**Step 3.3: User Confirmation**
+
+```javascript
+Options:
+- 确认执行: 开始应用所有修改
+- 调整计划: 重新回答问题调整范围
+- 取消操作: 放弃本次重规划
+```
+
+**Output**: Modification plan confirmed or adjusted
+
+---
+
+### Phase 4: Backup Creation
+
+**Process**:
+
+1. **Create Backup Directory**:
+```bash
+timestamp=$(date -u +"%Y-%m-%dT%H-%M-%S")
+backup_dir=".workflow/active/$SESSION_ID/.process/backup/replan-$timestamp"
+mkdir -p "$backup_dir"
+```
+
+2. **Backup All Affected Files**:
+   - IMPL_PLAN.md
+   - TODO_LIST.md
+   - workflow-session.json
+   - Affected task JSONs
+
+3. **Create Backup Manifest**:
+```markdown
+# Replan Backup Manifest
+
+**Timestamp**: {timestamp}
+**Reason**: {replan_reason}
+**Scope**: {modification_scope}
+
+## Restoration Command
+cp {backup_dir}/* .workflow/active/{session}/
+```
+
+**Output**: All files safely backed up with manifest
+
+---
+
+### Phase 5: Apply Modifications
+
+**Step 5.1: Update IMPL_PLAN.md** (if needed)
+
+Use Edit tool to modify specific sections:
+- Update affected technical sections
+- Update modification date
+
+**Step 5.2: Update TODO_LIST.md** (if needed)
+
+- Add new tasks with `[ ]` checkbox
+- Mark deleted tasks as `[x] ~~task~~ (已废弃)`
+- Update modified task descriptions
+
+**Step 5.3: Update Task JSONs**
+
+For each affected task:
+```typescript
+const updated_task = {
+  ...task,
+  context: {
+    ...task.context,
+    requirements: [...updated_requirements],
+    acceptance: [...updated_acceptance]
+  },
+  flow_control: {
+    ...task.flow_control,
+    implementation_approach: [...updated_steps]
+  }
+};
+
+Write({
+  file_path: `.workflow/active/${SESSION_ID}/.task/${task_id}.json`,
+  content: JSON.stringify(updated_task, null, 2)
+});
+```
+
+**Step 5.4: Create New Tasks** (if needed)
+
+Generate complete task JSON with all required fields:
+- id, title, status
+- meta (type, agent)
+- context (requirements, focus_paths, acceptance)
+- flow_control (pre_analysis, implementation_approach, target_files)
+
+**Step 5.5: Delete Obsolete Tasks** (if needed)
+
+Move to backup instead of hard delete:
+```bash
+mv ".workflow/active/$SESSION_ID/.task/{task-id}.json" "$backup_dir/"
+```
+
+**Step 5.6: Update Session Metadata**
+
+Update workflow-session.json:
+- progress.current_tasks
+- progress.last_replan
+- replan_history array
+
+**Output**: All modifications applied, artifacts updated
+
+---
+
+### Phase 6: Verification & Summary
+
+**Step 6.1: Verify Consistency**
+
+1. Validate all task JSONs are valid JSON
+2. Check task count within limits (max 10)
+3. Verify dependency graph is acyclic
+
+**Step 6.2: Generate Change Summary**
+
+```markdown
+## 重规划完成
+
+### 会话信息
+- **Session**: {session-id}
+- **时间**: {timestamp}
+- **备份**: {backup-path}
+
+### 变更摘要
+**范围**: {scope}
+**原因**: {reason}
+
+### 修改的文件
+- ✓ IMPL_PLAN.md: {changes}
+- ✓ TODO_LIST.md: {changes}
+- ✓ Task JSONs: {count} files updated
+
+### 任务变更
+- **新增**: {task-ids}
+- **删除**: {task-ids}
+- **更新**: {task-ids}
+
+### 回滚方法
+cp {backup-path}/* .workflow/active/{session}/
+```
+
+**Output**: Summary displayed, replan complete
+
+---
+
+## TodoWrite Progress Tracking
+
+### Session Mode Progress
+
+```json
+[
+  {"content": "检测模式和发现会话", "status": "completed", "activeForm": "检测模式和发现会话"},
+  {"content": "交互式需求明确", "status": "completed", "activeForm": "交互式需求明确"},
+  {"content": "影响分析和计划生成", "status": "completed", "activeForm": "影响分析和计划生成"},
+  {"content": "创建备份", "status": "completed", "activeForm": "创建备份"},
+  {"content": "更新会话产出文件", "status": "completed", "activeForm": "更新会话产出文件"},
+  {"content": "验证一致性", "status": "completed", "activeForm": "验证一致性"}
+]
+```
+
+### Task Mode Progress
+
+```json
+[
+  {"content": "检测会话和加载任务", "status": "completed", "activeForm": "检测会话和加载任务"},
+  {"content": "交互式更新确认", "status": "completed", "activeForm": "交互式更新确认"},
+  {"content": "应用任务修改", "status": "completed", "activeForm": "应用任务修改"}
+]
+```
+
+## Error Handling
+
+### Session Errors
+
+```bash
+# No active session found
+ERROR: No active session found
+Run /workflow:session:start to create a session
+
+# Session not found
+ERROR: Session WFS-invalid not found
+Available sessions: [list]
+
+# No changes specified
+WARNING: No modifications specified
+Use --interactive mode or provide requirements
+```
+
+### Task Errors
+
+```bash
+# Task not found
+ERROR: Task IMPL-999 not found in session
+Available tasks: [list]
+
+# Task completed
+WARNING: Task IMPL-001 is completed
+Consider creating new task for additional work
+
+# Circular dependency
+ERROR: Circular dependency detected
+Resolve dependency conflicts before proceeding
+```
+
+### Validation Errors
+
+```bash
+# Task limit exceeded
+ERROR: Replan would create 12 tasks (limit: 10)
+Consider: combining tasks, splitting sessions, or removing tasks
+
+# Invalid JSON
+ERROR: Generated invalid JSON
+Backup preserved, rolling back changes
+```
+
+## File Structure
+
+```
+.workflow/active/WFS-session-name/
+├── workflow-session.json
+├── IMPL_PLAN.md
+├── TODO_LIST.md
+├── .task/
+│   ├── IMPL-001.json
+│   ├── IMPL-002.json
+│   └── IMPL-003.json
+└── .process/
+    ├── context-package.json
+    └── backup/
+        └── replan-{timestamp}/
+            ├── MANIFEST.md
+            ├── IMPL_PLAN.md
+            ├── TODO_LIST.md
+            ├── workflow-session.json
+            └── IMPL-*.json
+```
+
+## Examples
+
+### Session Replan - Add Feature
+
+```bash
+/workflow:replan "添加双因素认证支持"
+
+# Interactive clarification
+Q: 修改范围?
+A: 全面重规划
+
+Q: 受影响模块?
+A: 认证模块, API接口
+
+Q: 任务变更?
+A: 添加新任务, 更新内容
+
+# Execution
+✓ 创建备份
+✓ 更新 IMPL_PLAN.md
+✓ 更新 TODO_LIST.md
+✓ 创建 IMPL-004.json
+✓ 更新 IMPL-001.json, IMPL-002.json
+
+重规划完成! 新增 1 任务,更新 2 任务
+```
+
+### Task Replan - Update Requirements
+
+```bash
+/workflow:replan IMPL-001 "支持 OAuth2.0 标准"
+
+# Interactive clarification
+Q: 更新部分?
+A: 需求和验收标准, 实现方案
+
+Q: 影响其他任务?
+A: 是,需要同步更新依赖任务
+
+# Execution
+✓ 创建备份
+✓ 更新 IMPL-001.json
+✓ 更新 IMPL-002.json (依赖任务)
+
+任务重规划完成! 更新 2 个任务
+```

.claude/commands/workflow/resume.md

@@ -1,105 +0,0 @@
----
-name: resume
-description: Intelligent workflow session resumption with automatic progress analysis
-argument-hint: "session-id for workflow session to resume"
-allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
----
-
-# Sequential Workflow Resume Command
-
-## Usage
-```bash
-/workflow:resume "<session-id>"
-```
-
-## Purpose
-**Sequential command coordination for workflow resumption** by first analyzing current session status, then continuing execution with special resume context. This command orchestrates intelligent session resumption through two-step process.
-
-## Command Coordination Workflow
-
-### Phase 1: Status Analysis
-1. **Call status command**: Execute `/workflow:status` to analyze current session state
-2. **Verify session information**: Check session ID, progress, and current task status
-3. **Identify resume point**: Determine where workflow was interrupted
-
-### Phase 2: Resume Execution
-1. **Call execute with resume flag**: Execute `/workflow:execute --resume-session="{session-id}"`
-2. **Pass session context**: Provide analyzed session information to execute command
-3. **Direct agent execution**: Skip discovery phase, directly enter TodoWrite and agent execution
-
-## Implementation Protocol
-
-### Sequential Command Execution
-```bash
-# Phase 1: Analyze current session status
-SlashCommand(command="/workflow:status")
-
-# Phase 2: Resume execution with special flag
-SlashCommand(command="/workflow:execute --resume-session=\"{session-id}\"")
-```
-
-### Progress Tracking
-```javascript
-TodoWrite({
-  todos: [
-    {
-      content: "Analyze current session status and progress",
-      status: "in_progress",
-      activeForm: "Analyzing session status"
-    },
-    {
-      content: "Resume workflow execution with session context",
-      status: "pending",
-      activeForm: "Resuming workflow execution"
-    }
-  ]
-});
-```
-
-## Resume Information Flow
-
-### Status Analysis Results
-The `/workflow:status` command provides:
-- **Session ID**: Current active session identifier
-- **Current Progress**: Completed, in-progress, and pending tasks
-- **Interruption Point**: Last executed task and next pending task
-- **Session State**: Overall workflow status
-
-### Execute Command Context
-The special `--resume-session` flag tells `/workflow:execute`:
-- **Skip Discovery**: Don't search for sessions, use provided session ID
-- **Direct Execution**: Go straight to TodoWrite generation and agent launching
-- **Context Restoration**: Use existing session state and summaries
-- **Resume Point**: Continue from identified interruption point
-
-## Error Handling
-
-### Session Validation Failures
-- **Session not found**: Report missing session, suggest available sessions
-- **Session inactive**: Recommend activating session first
-- **Status command fails**: Retry once, then report analysis failure
-
-### Execute Resumption Failures
-- **No pending tasks**: Report workflow completion status
-- **Execute command fails**: Report resumption failure, suggest manual intervention
-
-## Success Criteria
-1. **Status analysis complete**: Session state properly analyzed and reported
-2. **Execute command launched**: Resume execution started with proper context
-3. **Agent coordination**: TodoWrite and agent execution initiated successfully
-4. **Context preservation**: Session state and progress properly maintained
-
-## Related Commands
-
-**Prerequisite Commands**:
-- `/workflow:plan` or `/workflow:execute` - Workflow must be in progress or paused
-
-**Called by This Command** (2 phases):
-- `/workflow:status` - Phase 1: Analyze current session status and identify resume point
-- `/workflow:execute` - Phase 2: Resume execution with `--resume-session` flag
-
-**Follow-up Commands**:
-- None - Workflow continues automatically via `/workflow:execute`
-
----
-*Sequential command coordination for workflow session resumption*

.claude/commands/workflow/review-fix.md

@@ -0,0 +1,606 @@
+---
+name: review-fix
+description: Automated fixing of code review findings with AI-powered planning and coordinated execution. Uses intelligent grouping, multi-stage timeline coordination, and test-driven verification.
+argument-hint: "<export-file|review-dir> [--resume] [--max-iterations=N]"
+allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*), Task(*), Edit(*), Write(*)
+---
+
+# Workflow Review-Fix Command
+
+## Quick Start
+
+```bash
+# Fix from exported findings file (session-based path)
+/workflow:review-fix .workflow/active/WFS-123/.review/fix-export-1706184622000.json
+
+# Fix from review directory (auto-discovers latest export)
+/workflow:review-fix .workflow/active/WFS-123/.review/
+
+# Resume interrupted fix session
+/workflow:review-fix --resume
+
+# Custom max retry attempts per finding
+/workflow:review-fix .workflow/active/WFS-123/.review/ --max-iterations=5
+```
+
+**Fix Source**: Exported findings from review cycle dashboard
+**Output Directory**: `{review-dir}/fixes/{fix-session-id}/` (within session .review/)
+**Default Max Iterations**: 3 (per finding, adjustable)
+**CLI Tools**: @cli-planning-agent (planning), @cli-execute-agent (fixing)
+
+## What & Why
+
+### Core Concept
+Automated fix orchestrator with **two-phase architecture**: AI-powered planning followed by coordinated parallel/serial execution. Generates fix timeline with intelligent grouping and dependency analysis, then executes fixes with conservative test verification.
+
+**Fix Process**:
+- **Planning Phase**: AI analyzes findings, generates fix plan with grouping and execution strategy
+- **Execution Phase**: Main orchestrator coordinates agents per timeline stages
+- **No rigid structure**: Adapts to task requirements, not bound to fixed JSON format
+
+**vs Manual Fixing**:
+- **Manual**: Developer reviews findings one-by-one, fixes sequentially
+- **Automated**: AI groups related issues, executes in optimal parallel/serial order with automatic test verification
+
+### Value Proposition
+1. **Intelligent Planning**: AI-powered analysis identifies optimal grouping and execution strategy
+2. **Multi-stage Coordination**: Supports complex parallel + serial execution with dependency management
+3. **Conservative Safety**: Mandatory test verification with automatic rollback on failure
+4. **Resume Support**: Checkpoint-based recovery for interrupted sessions
+
+### Orchestrator Boundary (CRITICAL)
+- **ONLY command** for automated review finding fixes
+- Manages: Planning phase coordination, stage-based execution, agent scheduling, progress tracking
+- Delegates: Fix planning to @cli-planning-agent, fix execution to @cli-execute-agent
+
+
+### Execution Flow
+
+```
+Phase 1: Discovery & Initialization
+   └─ Validate export file, create fix session structure, initialize state files
+
+Phase 2: Planning Coordination (@cli-planning-agent)
+   ├─ Analyze findings for patterns and dependencies
+   ├─ Group by file + dimension + root cause similarity
+   ├─ Determine execution strategy (parallel/serial/hybrid)
+   ├─ Generate fix timeline with stages
+   └─ Output: fix-plan.json
+
+Phase 3: Execution Orchestration (Stage-based)
+   For each timeline stage:
+   ├─ Load groups for this stage
+   ├─ If parallel: Launch all group agents simultaneously
+   ├─ If serial: Execute groups sequentially
+   ├─ Each agent:
+   │  ├─ Analyze code context
+   │  ├─ Apply fix per strategy
+   │  ├─ Run affected tests
+   │  ├─ On test failure: Rollback, retry up to max_iterations
+   │  └─ On success: Commit, update fix-progress-{N}.json
+   └─ Advance to next stage
+
+Phase 4: Completion & Aggregation
+   └─ Aggregate results → Generate fix-summary.md → Update history → Output summary
+
+Phase 5: Session Completion (Optional)
+   └─ If all fixes successful → Prompt to complete workflow session
+```
+
+### Agent Roles
+
+| Agent | Responsibility |
+|-------|---------------|
+| **Orchestrator** | Input validation, session management, planning coordination, stage-based execution scheduling, progress tracking, aggregation |
+| **@cli-planning-agent** | Findings analysis, intelligent grouping (file+dimension+root cause), execution strategy determination (parallel/serial/hybrid), timeline generation with dependency mapping |
+| **@cli-execute-agent** | Fix execution per group, code context analysis, Edit tool operations, test verification, git rollback on failure, completion JSON generation |
+
+## Enhanced Features
+
+### 1. Two-Phase Architecture
+
+**Phase Separation**:
+
+| Phase | Agent | Output | Purpose |
+|-------|-------|--------|---------|
+| **Planning** | @cli-planning-agent | fix-plan.json | Analyze findings, group intelligently, determine optimal execution strategy |
+| **Execution** | @cli-execute-agent | completions/*.json | Execute fixes per plan with test verification and rollback |
+
+**Benefits**:
+- Clear separation of concerns (analysis vs execution)
+- Reusable plans (can re-execute without re-planning)
+- Better error isolation (planning failures vs execution failures)
+
+### 2. Intelligent Grouping Strategy
+
+**Three-Level Grouping**:
+
+```javascript
+// Level 1: Primary grouping by file + dimension
+{file: "auth.ts", dimension: "security"} → Group A
+{file: "auth.ts", dimension: "quality"} → Group B
+{file: "query-builder.ts", dimension: "security"} → Group C
+
+// Level 2: Secondary grouping by root cause similarity
+Group A findings → Semantic similarity analysis (threshold 0.7)
+  → Sub-group A1: "missing-input-validation" (findings 1, 2)
+  → Sub-group A2: "insecure-crypto" (finding 3)
+
+// Level 3: Dependency analysis
+Sub-group A1 creates validation utilities
+Sub-group C4 depends on those utilities
+→ A1 must execute before C4 (serial stage dependency)
+```
+
+**Similarity Computation**:
+- Combine: `description + recommendation + category`
+- Vectorize: TF-IDF or LLM embedding
+- Cluster: Greedy algorithm with cosine similarity > 0.7
+
+### 3. Execution Strategy Determination
+
+**Strategy Types**:
+
+| Strategy | When to Use | Stage Structure |
+|----------|-------------|-----------------|
+| **Parallel** | All groups independent, different files | Single stage, all groups in parallel |
+| **Serial** | Strong dependencies, shared resources | Multiple stages, one group per stage |
+| **Hybrid** | Mixed dependencies | Multiple stages, parallel within stages |
+
+**Dependency Detection**:
+- Shared file modifications
+- Utility creation + usage patterns
+- Test dependency chains
+- Risk level clustering (high-risk groups isolated)
+
+### 4. Conservative Test Verification
+
+**Test Strategy** (per fix):
+
+```javascript
+// 1. Identify affected tests
+const testPattern = identifyTestPattern(finding.file);
+// e.g., "tests/auth/**/*.test.*" for src/auth/service.ts
+
+// 2. Run tests
+const result = await runTests(testPattern);
+
+// 3. Evaluate
+if (result.passRate < 100%) {
+  // Rollback
+  await gitCheckout(finding.file);
+
+  // Retry with failure context
+  if (attempts < maxIterations) {
+    const fixContext = analyzeFailure(result.stderr);
+    regenerateFix(finding, fixContext);
+    retry();
+  } else {
+    markFailed(finding.id);
+  }
+} else {
+  // Commit
+  await gitCommit(`Fix: ${finding.title} [${finding.id}]`);
+  markFixed(finding.id);
+}
+```
+
+**Pass Criteria**: 100% test pass rate (no partial fixes)
+
+## Core Responsibilities
+
+### Orchestrator
+
+**Phase 1: Discovery & Initialization**
+- Input validation: Check export file exists and is valid JSON
+- Auto-discovery: If review-dir provided, find latest `*-fix-export.json`
+- Session creation: Generate fix-session-id (`fix-{timestamp}`)
+- Directory structure: Create `{review-dir}/fixes/{fix-session-id}/` with subdirectories
+- State files: Initialize active-fix-session.json (session marker)
+- TodoWrite initialization: Set up 4-phase tracking
+
+**Phase 2: Planning Coordination**
+- Launch @cli-planning-agent with findings data and project context
+- Validate fix-plan.json output (schema conformance, includes metadata with session status)
+- Load plan into memory for execution phase
+- TodoWrite update: Mark planning complete, start execution
+
+**Phase 3: Execution Orchestration**
+- Load fix-plan.json timeline stages
+- For each stage:
+  - If parallel mode: Launch all group agents via `Promise.all()`
+  - If serial mode: Execute groups sequentially with `await`
+  - Assign agent IDs (agents update their fix-progress-{N}.json)
+- Handle agent failures gracefully (mark group as failed, continue)
+- Advance to next stage only when current stage complete
+
+**Phase 4: Completion & Aggregation**
+- Collect final status from all fix-progress-{N}.json files
+- Generate fix-summary.md with timeline and results
+- Update fix-history.json with new session entry
+- Remove active-fix-session.json
+- TodoWrite completion: Mark all phases done
+- Output summary to user
+
+**Phase 5: Session Completion (Optional)**
+- If all findings fixed successfully (no failures):
+  - Prompt user: "All fixes complete. Complete workflow session? [Y/n]"
+  - If confirmed: Execute `/workflow:session:complete` to archive session with lessons learned
+- If partial success (some failures):
+  - Output: "Some findings failed. Review fix-summary.md before completing session."
+  - Do NOT auto-complete session
+
+### Output File Structure
+
+```
+.workflow/active/WFS-{session-id}/.review/
+├── fix-export-{timestamp}.json     # Exported findings (input)
+└── fixes/{fix-session-id}/
+    ├── fix-plan.json               # Planning agent output (execution plan with metadata)
+    ├── fix-progress-1.json         # Group 1 progress (planning agent init → agent updates)
+    ├── fix-progress-2.json         # Group 2 progress (planning agent init → agent updates)
+    ├── fix-progress-3.json         # Group 3 progress (planning agent init → agent updates)
+    ├── fix-summary.md              # Final report (orchestrator generates)
+    ├── active-fix-session.json     # Active session marker
+    └── fix-history.json            # All sessions history
+```
+
+**File Producers**:
+- **Planning Agent**: `fix-plan.json` (with metadata), all `fix-progress-*.json` (initial state)
+- **Execution Agents**: Update assigned `fix-progress-{N}.json` in real-time
+
+
+### Agent Invocation Template
+
+**Planning Agent**:
+```javascript
+Task({
+  subagent_type: "cli-planning-agent",
+  description: `Generate fix plan and initialize progress files for ${findings.length} findings`,
+  prompt: `
+## Task Objective
+Analyze ${findings.length} code review findings and generate execution plan with intelligent grouping and timeline coordination.
+
+## Input Data
+Review Session: ${reviewId}
+Fix Session ID: ${fixSessionId}
+Total Findings: ${findings.length}
+
+Findings:
+${JSON.stringify(findings, null, 2)}
+
+Project Context:
+- Structure: ${projectStructure}
+- Test Framework: ${testFramework}
+- Git Status: ${gitStatus}
+
+## Output Requirements
+
+### 1. fix-plan.json
+Execute: cat ~/.claude/workflows/cli-templates/fix-plan-template.json
+
+Generate execution plan following template structure:
+
+**Key Generation Rules**:
+- **Metadata**: Populate fix_session_id, review_session_id, status ("planning"), created_at, started_at timestamps
+- **Execution Strategy**: Choose approach (parallel/serial/hybrid) based on dependency analysis, set parallel_limit and stages count
+- **Groups**: Create groups (G1, G2, ...) with intelligent grouping (see Analysis Requirements below), assign progress files (fix-progress-1.json, ...), populate fix_strategy with approach/complexity/test_pattern, assess risks, identify dependencies
+- **Timeline**: Define stages respecting dependencies, set execution_mode per stage, map groups to stages, calculate critical path
+
+### 2. fix-progress-{N}.json (one per group)
+Execute: cat ~/.claude/workflows/cli-templates/fix-progress-template.json
+
+For each group (G1, G2, G3, ...), generate fix-progress-{N}.json following template structure:
+
+**Initial State Requirements**:
+- Status: "pending", phase: "waiting"
+- Timestamps: Set last_update to now, others null
+- Findings: Populate from review findings with status "pending", all operation fields null
+- Summary: Initialize all counters to zero
+- Flow control: Empty implementation_approach array
+- Errors: Empty array
+
+**CRITICAL**: Ensure complete template structure - all fields must be present.
+
+## Analysis Requirements
+
+### Intelligent Grouping Strategy
+Group findings using these criteria (in priority order):
+
+1. **File Proximity**: Findings in same file or related files
+2. **Dimension Affinity**: Same dimension (security, performance, etc.)
+3. **Root Cause Similarity**: Similar underlying issues
+4. **Fix Approach Commonality**: Can be fixed with similar approach
+
+**Grouping Guidelines**:
+- Optimal group size: 2-5 findings per group
+- Avoid cross-cutting concerns in same group
+- Consider test isolation (different test suites → different groups)
+- Balance workload across groups for parallel execution
+
+### Execution Strategy Determination
+
+**Parallel Mode**: Use when groups are independent, no shared files
+**Serial Mode**: Use when groups have dependencies or shared resources
+**Hybrid Mode**: Use for mixed dependency graphs (recommended for most cases)
+
+**Dependency Analysis**:
+- Identify shared files between groups
+- Detect test dependency chains
+- Evaluate risk of concurrent modifications
+
+### Risk Assessment
+
+For each group, evaluate:
+- **Complexity**: Based on code structure, file size, existing tests
+- **Impact Scope**: Number of files affected, API surface changes
+- **Rollback Feasibility**: Ease of reverting changes if tests fail
+
+### Test Strategy
+
+For each group, determine:
+- **Test Pattern**: Glob pattern matching affected tests
+- **Pass Criteria**: All tests must pass (100% pass rate)
+- **Test Command**: Infer from project (package.json, pytest.ini, etc.)
+
+## Output Files
+
+Write to ${sessionDir}:
+- ./fix-plan.json
+- ./fix-progress-1.json
+- ./fix-progress-2.json
+- ./fix-progress-{N}.json (as many as groups created)
+
+## Quality Checklist
+
+Before finalizing outputs:
+- ✅ All findings assigned to exactly one group
+- ✅ Group dependencies correctly identified
+- ✅ Timeline stages respect dependencies
+- ✅ All progress files have complete initial structure
+- ✅ Test patterns are valid and specific
+- ✅ Risk assessments are realistic
+- ✅ Estimated times are reasonable
+  `
+})
+```
+
+**Execution Agent** (per group):
+```javascript
+Task({
+  subagent_type: "cli-execute-agent",
+  description: `Fix ${group.findings.length} issues: ${group.group_name}`,
+  prompt: `
+## Task Objective
+Execute fixes for code review findings in group ${group.group_id}. Update progress file in real-time with flow control tracking.
+
+## Assignment
+- Group ID: ${group.group_id}
+- Group Name: ${group.group_name}
+- Progress File: ${sessionDir}/${group.progress_file}
+- Findings Count: ${group.findings.length}
+- Max Iterations: ${maxIterations} (per finding)
+
+## Fix Strategy
+${JSON.stringify(group.fix_strategy, null, 2)}
+
+## Risk Assessment
+${JSON.stringify(group.risk_assessment, null, 2)}
+
+## Execution Flow
+
+### Initialization (Before Starting)
+
+1. Read ${group.progress_file} to load initial state
+2. Update progress file:
+   - assigned_agent: "${agentId}"
+   - status: "in-progress"
+   - started_at: Current ISO 8601 timestamp
+   - last_update: Current ISO 8601 timestamp
+3. Write updated state back to ${group.progress_file}
+
+### Main Execution Loop
+
+For EACH finding in ${group.progress_file}.findings:
+
+#### Step 1: Analyze Context
+
+**Before Step**:
+- Update finding: status→"in-progress", started_at→now()
+- Update current_finding: Populate with finding details, status→"analyzing", action→"Reading file and understanding code structure"
+- Update phase→"analyzing"
+- Update flow_control: Add "analyze_context" step to implementation_approach (status→"in-progress"), set current_step→"analyze_context"
+- Update last_update→now(), write to ${group.progress_file}
+
+**Action**:
+- Read file: finding.file
+- Understand code structure around line: finding.line
+- Analyze surrounding context (imports, dependencies, related functions)
+- Review recommendations: finding.recommendations
+
+**After Step**:
+- Update flow_control: Mark "analyze_context" step as "completed" with completed_at→now()
+- Update last_update→now(), write to ${group.progress_file}
+
+#### Step 2: Apply Fix
+
+**Before Step**:
+- Update current_finding: status→"fixing", action→"Applying code changes per recommendations"
+- Update phase→"fixing"
+- Update flow_control: Add "apply_fix" step to implementation_approach (status→"in-progress"), set current_step→"apply_fix"
+- Update last_update→now(), write to ${group.progress_file}
+
+**Action**:
+- Use Edit tool to implement code changes per finding.recommendations
+- Follow fix_strategy.approach
+- Maintain code style and existing patterns
+
+**After Step**:
+- Update flow_control: Mark "apply_fix" step as "completed" with completed_at→now()
+- Update last_update→now(), write to ${group.progress_file}
+
+#### Step 3: Test Verification
+
+**Before Step**:
+- Update current_finding: status→"testing", action→"Running test suite to verify fix"
+- Update phase→"testing"
+- Update flow_control: Add "run_tests" step to implementation_approach (status→"in-progress"), set current_step→"run_tests"
+- Update last_update→now(), write to ${group.progress_file}
+
+**Action**:
+- Run tests using fix_strategy.test_pattern
+- Require 100% pass rate
+- Capture test output
+
+**On Test Failure**:
+- Git rollback: \`git checkout -- \${finding.file}\`
+- Increment finding.attempts
+- Update flow_control: Mark "run_tests" step as "failed" with completed_at→now()
+- Update errors: Add entry (finding_id, error_type→"test_failure", message, timestamp)
+- If finding.attempts < ${maxIterations}:
+  - Reset flow_control: implementation_approach→[], current_step→null
+  - Retry from Step 1
+- Else:
+  - Update finding: status→"completed", result→"failed", error_message→"Max iterations reached", completed_at→now()
+  - Update summary counts, move to next finding
+
+**On Test Success**:
+- Update flow_control: Mark "run_tests" step as "completed" with completed_at→now()
+- Update last_update→now(), write to ${group.progress_file}
+- Proceed to Step 4
+
+#### Step 4: Commit Changes
+
+**Before Step**:
+- Update current_finding: status→"committing", action→"Creating git commit for successful fix"
+- Update phase→"committing"
+- Update flow_control: Add "commit_changes" step to implementation_approach (status→"in-progress"), set current_step→"commit_changes"
+- Update last_update→now(), write to ${group.progress_file}
+
+**Action**:
+- Git commit: \`git commit -m "fix(${finding.dimension}): ${finding.title} [${finding.id}]"\`
+- Capture commit hash
+
+**After Step**:
+- Update finding: status→"completed", result→"fixed", commit_hash→<captured>, test_passed→true, completed_at→now()
+- Update flow_control: Mark "commit_changes" step as "completed" with completed_at→now()
+- Update last_update→now(), write to ${group.progress_file}
+
+#### After Each Finding
+
+- Update summary: Recalculate counts (pending/in_progress/fixed/failed) and percent_complete
+- If all findings completed: Clear current_finding, reset flow_control
+- Update last_update→now(), write to ${group.progress_file}
+
+### Final Completion
+
+When all findings processed:
+- Update status→"completed", phase→"done", summary.percent_complete→100.0
+- Update last_update→now(), write final state to ${group.progress_file}
+
+## Critical Requirements
+
+### Progress File Updates
+- **MUST update after every significant action** (before/after each step)
+- **Always maintain complete structure** - never write partial updates
+- **Use ISO 8601 timestamps** - e.g., "2025-01-25T14:36:00Z"
+
+### Flow Control Format
+Follow action-planning-agent flow_control.implementation_approach format:
+- step: Identifier (e.g., "analyze_context", "apply_fix")
+- action: Human-readable description
+- status: "pending" | "in-progress" | "completed" | "failed"
+- started_at: ISO 8601 timestamp or null
+- completed_at: ISO 8601 timestamp or null
+
+### Error Handling
+- Capture all errors in errors[] array
+- Never leave progress file in invalid state
+- Always write complete updates, never partial
+- On unrecoverable error: Mark group as failed, preserve state
+
+## Test Patterns
+Use fix_strategy.test_pattern to run affected tests:
+- Pattern: ${group.fix_strategy.test_pattern}
+- Command: Infer from project (npm test, pytest, etc.)
+- Pass Criteria: 100% pass rate required
+  `
+})
+```
+
+
+
+### Error Handling
+
+**Planning Failures**:
+- Invalid template → Abort with error message
+- Insufficient findings data → Request complete export
+- Planning timeout → Retry once, then fail gracefully
+
+**Execution Failures**:
+- Agent crash → Mark group as failed, continue with other groups
+- Test command not found → Skip test verification, warn user
+- Git operations fail → Abort with error, preserve state
+
+**Rollback Scenarios**:
+- Test failure after fix → Automatic `git checkout` rollback
+- Max iterations reached → Leave file unchanged, mark as failed
+- Unrecoverable error → Rollback entire group, save checkpoint
+
+### TodoWrite Structure
+
+**Initialization**:
+```javascript
+TodoWrite({
+  todos: [
+    {content: "Phase 1: Discovery & Initialization", status: "completed"},
+    {content: "Phase 2: Planning", status: "in_progress"},
+    {content: "Phase 3: Execution", status: "pending"},
+    {content: "Phase 4: Completion", status: "pending"}
+  ]
+});
+```
+
+**During Execution**:
+```javascript
+TodoWrite({
+  todos: [
+    {content: "Phase 1: Discovery & Initialization", status: "completed"},
+    {content: "Phase 2: Planning", status: "completed"},
+    {content: "Phase 3: Execution", status: "in_progress"},
+    {content: "  → Stage 1: Parallel execution (3 groups)", status: "completed"},
+    {content: "    • Group G1: Auth validation (2 findings)", status: "completed"},
+    {content: "    • Group G2: Query security (3 findings)", status: "completed"},
+    {content: "    • Group G3: Config quality (1 finding)", status: "completed"},
+    {content: "  → Stage 2: Serial execution (1 group)", status: "in_progress"},
+    {content: "    • Group G4: Dependent fixes (2 findings)", status: "in_progress"},
+    {content: "Phase 4: Completion", status: "pending"}
+  ]
+});
+```
+
+**Update Rules**:
+- Add stage items dynamically based on fix-plan.json timeline
+- Add group items per stage
+- Mark completed immediately after each group finishes
+- Update parent phase status when all child items complete
+
+## Best Practices
+
+1. **Trust AI Planning**: Planning agent's grouping and execution strategy are based on dependency analysis
+2. **Conservative Approach**: Test verification is mandatory - no fixes kept without passing tests
+3. **Parallel Efficiency**: Default 3 concurrent agents balances speed and resource usage
+4. **Resume Support**: Fix sessions can resume from checkpoints after interruption
+5. **Manual Review**: Always review failed fixes manually - may require architectural changes
+6. **Incremental Fixing**: Start with small batches (5-10 findings) before large-scale fixes
+
+## Related Commands
+
+### View Fix Progress
+Use `ccw view` to open the workflow dashboard in browser:
+
+```bash
+ccw view
+```
+
+

.claude/commands/workflow/review-module-cycle.md

@@ -0,0 +1,771 @@
+---
+name: review-module-cycle
+description: Independent multi-dimensional code review for specified modules/files. Analyzes specific code paths across 7 dimensions with hybrid parallel-iterative execution, independent of workflow sessions.
+argument-hint: "<path-pattern> [--dimensions=security,architecture,...] [--max-iterations=N]"
+allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*), Task(*)
+---
+
+# Workflow Review-Module-Cycle Command
+
+## Quick Start
+
+```bash
+# Review specific module (all 7 dimensions)
+/workflow:review-module-cycle src/auth/**
+
+# Review multiple modules
+/workflow:review-module-cycle src/auth/**,src/payment/**
+
+# Review with custom dimensions
+/workflow:review-module-cycle src/payment/** --dimensions=security,architecture,quality
+
+# Review specific files
+/workflow:review-module-cycle src/payment/processor.ts,src/payment/validator.ts
+```
+
+**Review Scope**: Specified modules/files only (independent of git history)
+**Session Requirement**: Auto-creates workflow session via `/workflow:session:start`
+**Output Directory**: `.workflow/active/WFS-{session-id}/.review/` (session-based)
+**Default Dimensions**: Security, Architecture, Quality, Action-Items, Performance, Maintainability, Best-Practices
+**Max Iterations**: 3 (adjustable via --max-iterations)
+**Default Iterations**: 1 (deep-dive runs once; use --max-iterations=0 to skip)
+**CLI Tools**: Gemini → Qwen → Codex (fallback chain)
+
+## What & Why
+
+### Core Concept
+Independent multi-dimensional code review orchestrator with **hybrid parallel-iterative execution** for comprehensive quality assessment of **specific modules or files**.
+
+**Review Scope**:
+- **Module-based**: Reviews specified file patterns (e.g., `src/auth/**`, `*.ts`)
+- **Session-integrated**: Runs within workflow session context for unified tracking
+- **Output location**: `.review/` subdirectory within active session
+
+**vs Session Review**:
+- **Session Review** (`review-session-cycle`): Reviews git changes within a workflow session
+- **Module Review** (`review-module-cycle`): Reviews any specified code paths, regardless of git history
+- **Common output**: Both use same `.review/` directory structure within session
+
+### Value Proposition
+1. **Module-Focused Review**: Target specific code areas independent of git history
+2. **Session-Integrated**: Review results tracked within workflow session for unified management
+3. **Comprehensive Coverage**: Same 7 specialized dimensions as session review
+4. **Intelligent Prioritization**: Automatic identification of critical issues and cross-cutting concerns
+5. **Unified Archive**: Review results archived with session for historical reference
+
+### Orchestrator Boundary (CRITICAL)
+- **ONLY command** for independent multi-dimensional module review
+- Manages: dimension coordination, aggregation, iteration control, progress tracking
+- Delegates: Code exploration and analysis to @cli-explore-agent, dimension-specific reviews via Deep Scan mode
+
+## How It Works
+
+### Execution Flow
+
+```
+Phase 1: Discovery & Initialization
+   └─ Resolve file patterns, validate paths, initialize state, create output structure
+
+Phase 2: Parallel Reviews (for each dimension)
+   ├─ Launch 7 review agents simultaneously
+   ├─ Each executes CLI analysis via Gemini/Qwen on specified files
+   ├─ Generate dimension JSON + markdown reports
+   └─ Update review-progress.json
+
+Phase 3: Aggregation
+   ├─ Load all dimension JSON files
+   ├─ Calculate severity distribution (critical/high/medium/low)
+   ├─ Identify cross-cutting concerns (files in 3+ dimensions)
+   └─ Decision:
+      ├─ Critical findings OR high > 5 OR critical files → Phase 4 (Iterate)
+      └─ Else → Phase 5 (Complete)
+
+Phase 4: Iterative Deep-Dive (optional)
+   ├─ Select critical findings (max 5 per iteration)
+   ├─ Launch deep-dive agents for root cause analysis
+   ├─ Generate remediation plans with impact assessment
+   ├─ Re-assess severity based on analysis
+   └─ Loop until no critical findings OR max iterations
+
+Phase 5: Completion
+   └─ Finalize review-progress.json
+```
+
+### Agent Roles
+
+| Agent | Responsibility |
+|-------|---------------|
+| **Orchestrator** | Phase control, path resolution, state management, aggregation logic, iteration control |
+| **@cli-explore-agent** (Review) | Execute dimension-specific code analysis via Deep Scan mode, generate findings JSON with dual-source strategy (Bash + Gemini), create structured analysis reports |
+| **@cli-explore-agent** (Deep-dive) | Focused root cause analysis using dependency mapping, remediation planning with architectural insights, impact assessment, severity re-assessment |
+
+## Enhanced Features
+
+### 1. Review Dimensions Configuration
+
+**7 Specialized Dimensions** with priority-based allocation:
+
+| Dimension | Template | Priority | Timeout |
+|-----------|----------|----------|---------|
+| **Security** | 03-assess-security-risks.txt | 1 (Critical) | 60min |
+| **Architecture** | 02-review-architecture.txt | 2 (High) | 60min |
+| **Quality** | 02-review-code-quality.txt | 3 (Medium) | 40min |
+| **Action-Items** | 02-analyze-code-patterns.txt | 2 (High) | 40min |
+| **Performance** | 03-analyze-performance.txt | 3 (Medium) | 60min |
+| **Maintainability** | 02-review-code-quality.txt* | 3 (Medium) | 40min |
+| **Best-Practices** | 03-review-quality-standards.txt | 3 (Medium) | 40min |
+
+*Custom focus: "Assess technical debt and maintainability"
+
+**Category Definitions by Dimension**:
+
+```javascript
+const CATEGORIES = {
+  security: ['injection', 'authentication', 'authorization', 'encryption', 'input-validation', 'access-control', 'data-exposure'],
+  architecture: ['coupling', 'cohesion', 'layering', 'dependency', 'pattern-violation', 'scalability', 'separation-of-concerns'],
+  quality: ['code-smell', 'duplication', 'complexity', 'naming', 'error-handling', 'testability', 'readability'],
+  'action-items': ['requirement-coverage', 'acceptance-criteria', 'documentation', 'deployment-readiness', 'missing-functionality'],
+  performance: ['n-plus-one', 'inefficient-query', 'memory-leak', 'blocking-operation', 'caching', 'resource-usage'],
+  maintainability: ['technical-debt', 'magic-number', 'long-method', 'large-class', 'dead-code', 'commented-code'],
+  'best-practices': ['convention-violation', 'anti-pattern', 'deprecated-api', 'missing-validation', 'inconsistent-style']
+};
+```
+
+### 2. Path Pattern Resolution
+
+**Syntax Rules**:
+- All paths are **relative** from project root (e.g., `src/auth/**` not `/src/auth/**`)
+- Multiple patterns: comma-separated, **no spaces** (e.g., `src/auth/**,src/payment/**`)
+- Glob and specific files can be mixed (e.g., `src/auth/**,src/config.ts`)
+
+**Supported Patterns**:
+| Pattern Type | Example | Description |
+|--------------|---------|-------------|
+| Glob directory | `src/auth/**` | All files under src/auth/ |
+| Glob with extension | `src/**/*.ts` | All .ts files under src/ |
+| Specific file | `src/payment/processor.ts` | Single file |
+| Multiple patterns | `src/auth/**,src/payment/**` | Comma-separated (no spaces) |
+
+**Resolution Process**:
+1. Parse input pattern (split by comma, trim whitespace)
+2. Expand glob patterns to file list via `find` command
+3. Validate all files exist and are readable
+4. Error if pattern matches 0 files
+5. Store resolved file list in review-state.json
+
+### 3. Aggregation Logic
+
+**Cross-Cutting Concern Detection**:
+1. Files appearing in 3+ dimensions = **Critical Files**
+2. Same issue pattern across dimensions = **Systemic Issue**
+3. Severity clustering in specific files = **Hotspots**
+
+**Deep-Dive Selection Criteria**:
+- All critical severity findings (priority 1)
+- Top 3 high-severity findings in critical files (priority 2)
+- Max 5 findings per iteration (prevent overwhelm)
+
+### 4. Severity Assessment
+
+**Severity Levels**:
+- **Critical**: Security vulnerabilities, data corruption risks, system-wide failures, authentication/authorization bypass
+- **High**: Feature degradation, performance bottlenecks, architecture violations, significant technical debt
+- **Medium**: Code smells, minor performance issues, style inconsistencies, maintainability concerns
+- **Low**: Documentation gaps, minor refactoring opportunities, cosmetic issues
+
+**Iteration Trigger**:
+- Critical findings > 0 OR
+- High findings > 5 OR
+- Critical files count > 0
+
+## Core Responsibilities
+
+### Orchestrator
+
+**Phase 1: Discovery & Initialization**
+
+**Step 1: Session Creation**
+```javascript
+// Create workflow session for this review (type: review)
+SlashCommand(command="/workflow:session:start --type review \"Code review for [target_pattern]\"")
+
+// Parse output
+const sessionId = output.match(/SESSION_ID: (WFS-[^\s]+)/)[1];
+```
+
+**Step 2: Path Resolution & Validation**
+```bash
+# Expand glob pattern to file list (relative paths from project root)
+find . -path "./src/auth/**" -type f | sed 's|^\./||'
+
+# Validate files exist and are readable
+for file in ${resolvedFiles[@]}; do
+  test -r "$file" || error "File not readable: $file"
+done
+```
+- Parse and expand file patterns (glob support): `src/auth/**` → actual file list
+- Validation: Ensure all specified files exist and are readable
+- Store as **relative paths** from project root (e.g., `src/auth/service.ts`)
+- Agents construct absolute paths dynamically during execution
+
+**Step 3: Output Directory Setup**
+- Output directory: `.workflow/active/${sessionId}/.review/`
+- Create directory structure:
+  ```bash
+  mkdir -p ${sessionDir}/.review/{dimensions,iterations,reports}
+  ```
+
+**Step 4: Initialize Review State**
+- State initialization: Create `review-state.json` with metadata, dimensions, max_iterations, resolved_files (merged metadata + state)
+- Progress tracking: Create `review-progress.json` for progress tracking
+
+**Step 5: TodoWrite Initialization**
+- Set up progress tracking with hierarchical structure
+- Mark Phase 1 completed, Phase 2 in_progress
+
+**Phase 2: Parallel Review Coordination**
+- Launch 7 @cli-explore-agent instances simultaneously (Deep Scan mode)
+- Pass dimension-specific context (template, timeout, custom focus, **target files**)
+- Monitor completion via review-progress.json updates
+- TodoWrite updates: Mark dimensions as completed
+- CLI tool fallback: Gemini → Qwen → Codex (on error/timeout)
+
+**Phase 3: Aggregation**
+- Load all dimension JSON files from dimensions/
+- Calculate severity distribution: Count by critical/high/medium/low
+- Identify cross-cutting concerns: Files in 3+ dimensions
+- Select deep-dive findings: Critical + high in critical files (max 5)
+- Decision logic: Iterate if critical > 0 OR high > 5 OR critical files exist
+- Update review-state.json with aggregation results
+
+**Phase 4: Iteration Control**
+- Check iteration count < max_iterations (default 3)
+- Launch deep-dive agents for selected findings
+- Collect remediation plans and re-assessed severities
+- Update severity distribution based on re-assessments
+- Record iteration in review-state.json
+- Loop back to aggregation if still have critical/high findings
+
+**Phase 5: Completion**
+- Finalize review-progress.json with completion statistics
+- Update review-state.json with completion_time and phase=complete
+- TodoWrite completion: Mark all tasks done
+
+
+
+### Output File Structure
+
+```
+.workflow/active/WFS-{session-id}/.review/
+├── review-state.json                    # Orchestrator state machine (includes metadata)
+├── review-progress.json                 # Real-time progress for dashboard
+├── dimensions/                          # Per-dimension results
+│   ├── security.json
+│   ├── architecture.json
+│   ├── quality.json
+│   ├── action-items.json
+│   ├── performance.json
+│   ├── maintainability.json
+│   └── best-practices.json
+├── iterations/                          # Deep-dive results
+│   ├── iteration-1-finding-{uuid}.json
+│   └── iteration-2-finding-{uuid}.json
+└── reports/                             # Human-readable reports
+    ├── security-analysis.md
+    ├── security-cli-output.txt
+    ├── deep-dive-1-{uuid}.md
+    └── ...
+```
+
+**Session Context**:
+```
+.workflow/active/WFS-{session-id}/
+├── workflow-session.json
+├── IMPL_PLAN.md
+├── TODO_LIST.md
+├── .task/
+├── .summaries/
+└── .review/                             # Review results (this command)
+    └── (structure above)
+```
+
+### Review State JSON
+
+**Purpose**: Unified state machine and metadata (merged from metadata + state)
+
+```json
+{
+  "review_id": "review-20250125-143022",
+  "review_type": "module",
+  "session_id": "WFS-auth-system",
+  "metadata": {
+    "created_at": "2025-01-25T14:30:22Z",
+    "target_pattern": "src/auth/**",
+    "resolved_files": [
+      "src/auth/service.ts",
+      "src/auth/validator.ts",
+      "src/auth/middleware.ts"
+    ],
+    "dimensions": ["security", "architecture", "quality", "action-items", "performance", "maintainability", "best-practices"],
+    "max_iterations": 3
+  },
+  "phase": "parallel|aggregate|iterate|complete",
+  "current_iteration": 1,
+  "dimensions_reviewed": ["security", "architecture", "quality", "action-items", "performance", "maintainability", "best-practices"],
+  "selected_strategy": "comprehensive",
+  "next_action": "execute_parallel_reviews|aggregate_findings|execute_deep_dive|generate_final_report|complete",
+  "severity_distribution": {
+    "critical": 2,
+    "high": 5,
+    "medium": 12,
+    "low": 8
+  },
+  "critical_files": [...],
+  "iterations": [...],
+  "completion_criteria": {...}
+}
+```
+
+### Review Progress JSON
+
+**Purpose**: Real-time dashboard updates via polling
+
+```json
+{
+  "review_id": "review-20250125-143022",
+  "last_update": "2025-01-25T14:35:10Z",
+  "phase": "parallel|aggregate|iterate|complete",
+  "current_iteration": 1,
+  "progress": {
+    "parallel_review": {
+      "total_dimensions": 7,
+      "completed": 5,
+      "in_progress": 2,
+      "percent_complete": 71
+    },
+    "deep_dive": {
+      "total_findings": 6,
+      "analyzed": 2,
+      "in_progress": 1,
+      "percent_complete": 33
+    }
+  },
+  "agent_status": [
+    {
+      "agent_type": "review-agent",
+      "dimension": "security",
+      "status": "completed",
+      "started_at": "2025-01-25T14:30:00Z",
+      "completed_at": "2025-01-25T15:15:00Z",
+      "duration_ms": 2700000
+    },
+    {
+      "agent_type": "deep-dive-agent",
+      "finding_id": "sec-001-uuid",
+      "status": "in_progress",
+      "started_at": "2025-01-25T14:32:00Z"
+    }
+  ],
+  "estimated_completion": "2025-01-25T16:00:00Z"
+}
+```
+
+### Agent Output Schemas
+
+**Agent-produced JSON files follow standardized schemas**:
+
+1. **Dimension Results** (cli-explore-agent output from parallel reviews)
+   - Schema: `~/.claude/workflows/cli-templates/schemas/review-dimension-results-schema.json`
+   - Output: `{output-dir}/dimensions/{dimension}.json`
+   - Contains: findings array, summary statistics, cross_references
+
+2. **Deep-Dive Results** (cli-explore-agent output from iterations)
+   - Schema: `~/.claude/workflows/cli-templates/schemas/review-deep-dive-results-schema.json`
+   - Output: `{output-dir}/iterations/iteration-{N}-finding-{uuid}.json`
+   - Contains: root_cause, remediation_plan, impact_assessment, reassessed_severity
+
+### Agent Invocation Template
+
+**Review Agent** (parallel execution, 7 instances):
+
+```javascript
+Task(
+  subagent_type="cli-explore-agent",
+  run_in_background=false,
+  description=`Execute ${dimension} review analysis via Deep Scan`,
+  prompt=`
+    ## Task Objective
+    Conduct comprehensive ${dimension} code exploration and analysis using Deep Scan mode (Bash + Gemini dual-source strategy) for specified module files
+
+    ## Analysis Mode Selection
+    Use **Deep Scan mode** for this review:
+    - Phase 1: Bash structural scan for standard patterns (classes, functions, imports)
+    - Phase 2: Gemini semantic analysis for design intent, non-standard patterns, ${dimension}-specific concerns
+    - Phase 3: Synthesis with attribution (bash-discovered vs gemini-discovered findings)
+
+    ## MANDATORY FIRST STEPS (Execute by Agent)
+    **You (cli-explore-agent) MUST execute these steps in order:**
+    1. Read review state: ${reviewStateJsonPath}
+    2. Get target files: Read resolved_files from review-state.json
+    3. Validate file access: bash(ls -la ${targetFiles.join(' ')})
+    4. Execute: cat ~/.claude/workflows/cli-templates/schemas/review-dimension-results-schema.json (get output schema reference)
+    5. Read: .workflow/project-tech.json (technology stack and architecture context)
+    6. Read: .workflow/project-guidelines.json (user-defined constraints and conventions to validate against)
+
+    ## Review Context
+    - Review Type: module (independent)
+    - Review Dimension: ${dimension}
+    - Review ID: ${reviewId}
+    - Target Pattern: ${targetPattern}
+    - Resolved Files: ${resolvedFiles.length} files
+    - Output Directory: ${outputDir}
+
+    ## CLI Configuration
+    - Tool Priority: gemini → qwen → codex (fallback chain)
+    - Custom Focus: ${customFocus || 'Standard dimension analysis'}
+    - Mode: analysis (READ-ONLY)
+    - Context Pattern: ${targetFiles.map(f => `@${f}`).join(' ')}
+
+    ## Expected Deliverables
+
+    **Schema Reference**: Schema obtained in MANDATORY FIRST STEPS step 4, follow schema exactly
+
+    1. Dimension Results JSON: ${outputDir}/dimensions/${dimension}.json
+
+       **⚠️ CRITICAL JSON STRUCTURE REQUIREMENTS**:
+
+       Root structure MUST be array: \`[{ ... }]\` NOT \`{ ... }\`
+
+       Required top-level fields:
+       - dimension, review_id, analysis_timestamp (NOT timestamp/analyzed_at)
+       - cli_tool_used (gemini|qwen|codex), model, analysis_duration_ms
+       - summary (FLAT structure), findings, cross_references
+
+       Summary MUST be FLAT (NOT nested by_severity):
+       \`{ "total_findings": N, "critical": N, "high": N, "medium": N, "low": N, "files_analyzed": N, "lines_reviewed": N }\`
+
+       Finding required fields:
+       - id: format \`{dim}-{seq}-{uuid8}\` e.g., \`sec-001-a1b2c3d4\` (lowercase)
+       - severity: lowercase only (critical|high|medium|low)
+       - snippet (NOT code_snippet), impact (NOT exploit_scenario)
+       - metadata, iteration (0), status (pending_remediation), cross_references
+
+    2. Analysis Report: ${outputDir}/reports/${dimension}-analysis.md
+       - Human-readable summary with recommendations
+       - Grouped by severity: critical → high → medium → low
+       - Include file:line references for all findings
+
+    3. CLI Output Log: ${outputDir}/reports/${dimension}-cli-output.txt
+       - Raw CLI tool output for debugging
+       - Include full analysis text
+
+    ## Dimension-Specific Guidance
+    ${getDimensionGuidance(dimension)}
+
+    ## Success Criteria
+    - [ ] Schema obtained via cat review-dimension-results-schema.json
+    - [ ] All target files analyzed for ${dimension} concerns
+    - [ ] All findings include file:line references with code snippets
+    - [ ] Severity assessment follows established criteria (see reference)
+    - [ ] Recommendations are actionable with code examples
+    - [ ] JSON output follows schema exactly
+    - [ ] Report is comprehensive and well-organized
+  `
+)
+```
+
+**Deep-Dive Agent** (iteration execution):
+
+```javascript
+Task(
+  subagent_type="cli-explore-agent",
+  run_in_background=false,
+  description=`Deep-dive analysis for critical finding: ${findingTitle} via Dependency Map + Deep Scan`,
+  prompt=`
+    ## Task Objective
+    Perform focused root cause analysis using Dependency Map mode (for impact analysis) + Deep Scan mode (for semantic understanding) to generate comprehensive remediation plan for critical ${dimension} issue
+
+    ## Analysis Mode Selection
+    Use **Dependency Map mode** first to understand dependencies:
+    - Build dependency graph around ${file} to identify affected components
+    - Detect circular dependencies or tight coupling related to this finding
+    - Calculate change risk scores for remediation impact
+
+    Then apply **Deep Scan mode** for semantic analysis:
+    - Understand design intent and architectural context
+    - Identify non-standard patterns or implicit dependencies
+    - Extract remediation insights from code structure
+
+    ## Finding Context
+    - Finding ID: ${findingId}
+    - Original Dimension: ${dimension}
+    - Title: ${findingTitle}
+    - File: ${file}:${line}
+    - Severity: ${severity}
+    - Category: ${category}
+    - Original Description: ${description}
+    - Iteration: ${iteration}
+
+    ## MANDATORY FIRST STEPS (Execute by Agent)
+    **You (cli-explore-agent) MUST execute these steps in order:**
+    1. Read original finding: ${dimensionJsonPath}
+    2. Read affected file: ${file}
+    3. Identify related code: bash(grep -r "import.*${basename(file)}" ${projectDir}/src --include="*.ts")
+    4. Read test files: bash(find ${projectDir}/tests -name "*${basename(file, '.ts')}*" -type f)
+    5. Execute: cat ~/.claude/workflows/cli-templates/schemas/review-deep-dive-results-schema.json (get output schema reference)
+    6. Read: .workflow/project-tech.json (technology stack and architecture context)
+    7. Read: .workflow/project-guidelines.json (user-defined constraints for remediation compliance)
+
+    ## CLI Configuration
+    - Tool Priority: gemini → qwen → codex
+    - Template: ~/.claude/workflows/cli-templates/prompts/analysis/01-diagnose-bug-root-cause.txt
+    - Mode: analysis (READ-ONLY)
+
+    ## Expected Deliverables
+
+    **Schema Reference**: Schema obtained in MANDATORY FIRST STEPS step 5, follow schema exactly
+
+    1. Deep-Dive Results JSON: ${outputDir}/iterations/iteration-${iteration}-finding-${findingId}.json
+
+       **⚠️ CRITICAL JSON STRUCTURE REQUIREMENTS**:
+
+       Root structure MUST be array: \`[{ ... }]\` NOT \`{ ... }\`
+
+       Required top-level fields:
+       - finding_id, dimension, iteration, analysis_timestamp
+       - cli_tool_used, model, analysis_duration_ms
+       - original_finding, root_cause, remediation_plan
+       - impact_assessment, reassessed_severity, confidence_score, cross_references
+
+       All nested objects must follow schema exactly - read schema for field names
+
+    2. Analysis Report: ${outputDir}/reports/deep-dive-${iteration}-${findingId}.md
+       - Detailed root cause analysis
+       - Step-by-step remediation plan
+       - Impact assessment and rollback strategy
+
+    ## Success Criteria
+    - [ ] Schema obtained via cat review-deep-dive-results-schema.json
+    - [ ] Root cause clearly identified with supporting evidence
+    - [ ] Remediation plan is step-by-step actionable with exact file:line references
+    - [ ] Each step includes specific commands and validation tests
+    - [ ] Impact fully assessed (files, tests, breaking changes, dependencies)
+    - [ ] Severity re-evaluation justified with evidence
+    - [ ] Confidence score accurately reflects certainty of analysis
+    - [ ] JSON output follows schema exactly
+    - [ ] References include project-specific and external documentation
+  `
+)
+```
+
+### Dimension Guidance Reference
+
+```javascript
+function getDimensionGuidance(dimension) {
+  const guidance = {
+    security: `
+      Focus Areas:
+      - Input validation and sanitization
+      - Authentication and authorization mechanisms
+      - Data encryption (at-rest and in-transit)
+      - SQL/NoSQL injection vulnerabilities
+      - XSS, CSRF, and other web vulnerabilities
+      - Sensitive data exposure
+      - Access control and privilege escalation
+
+      Severity Criteria:
+      - Critical: Authentication bypass, SQL injection, RCE, sensitive data exposure
+      - High: Missing authorization checks, weak encryption, exposed secrets
+      - Medium: Missing input validation, insecure defaults, weak password policies
+      - Low: Security headers missing, verbose error messages, outdated dependencies
+    `,
+    architecture: `
+      Focus Areas:
+      - Layering and separation of concerns
+      - Coupling and cohesion
+      - Design pattern adherence
+      - Dependency management
+      - Scalability and extensibility
+      - Module boundaries
+      - API design consistency
+
+      Severity Criteria:
+      - Critical: Circular dependencies, god objects, tight coupling across layers
+      - High: Violated architectural principles, scalability bottlenecks
+      - Medium: Missing abstractions, inconsistent patterns, suboptimal design
+      - Low: Minor coupling issues, documentation gaps, naming inconsistencies
+    `,
+    quality: `
+      Focus Areas:
+      - Code duplication
+      - Complexity (cyclomatic, cognitive)
+      - Naming conventions
+      - Error handling patterns
+      - Code readability
+      - Comment quality
+      - Dead code
+
+      Severity Criteria:
+      - Critical: Severe complexity (CC > 20), massive duplication (>50 lines)
+      - High: High complexity (CC > 10), significant duplication, poor error handling
+      - Medium: Moderate complexity (CC > 5), naming issues, code smells
+      - Low: Minor duplication, documentation gaps, cosmetic issues
+    `,
+    'action-items': `
+      Focus Areas:
+      - Requirements coverage verification
+      - Acceptance criteria met
+      - Documentation completeness
+      - Deployment readiness
+      - Missing functionality
+      - Test coverage gaps
+      - Configuration management
+
+      Severity Criteria:
+      - Critical: Core requirements not met, deployment blockers
+      - High: Significant functionality missing, acceptance criteria not met
+      - Medium: Minor requirements gaps, documentation incomplete
+      - Low: Nice-to-have features missing, minor documentation gaps
+    `,
+    performance: `
+      Focus Areas:
+      - N+1 query problems
+      - Inefficient algorithms (O(n²) where O(n log n) possible)
+      - Memory leaks
+      - Blocking operations on main thread
+      - Missing caching opportunities
+      - Resource usage (CPU, memory, network)
+      - Database query optimization
+
+      Severity Criteria:
+      - Critical: Memory leaks, O(n²) in hot path, blocking main thread
+      - High: N+1 queries, missing indexes, inefficient algorithms
+      - Medium: Suboptimal caching, unnecessary computations, lazy loading issues
+      - Low: Minor optimization opportunities, redundant operations
+    `,
+    maintainability: `
+      Focus Areas:
+      - Technical debt indicators
+      - Magic numbers and hardcoded values
+      - Long methods (>50 lines)
+      - Large classes (>500 lines)
+      - Dead code and commented code
+      - Code documentation
+      - Test coverage
+
+      Severity Criteria:
+      - Critical: Massive methods (>200 lines), severe technical debt blocking changes
+      - High: Large methods (>100 lines), significant dead code, undocumented complex logic
+      - Medium: Magic numbers, moderate technical debt, missing tests
+      - Low: Minor refactoring opportunities, cosmetic improvements
+    `,
+    'best-practices': `
+      Focus Areas:
+      - Framework conventions adherence
+      - Language idioms
+      - Anti-patterns
+      - Deprecated API usage
+      - Coding standards compliance
+      - Error handling patterns
+      - Logging and monitoring
+
+      Severity Criteria:
+      - Critical: Severe anti-patterns, deprecated APIs with security risks
+      - High: Major convention violations, poor error handling, missing logging
+      - Medium: Minor anti-patterns, style inconsistencies, suboptimal patterns
+      - Low: Cosmetic style issues, minor convention deviations
+    `
+  };
+
+  return guidance[dimension] || 'Standard code review analysis';
+}
+```
+
+### Completion Conditions
+
+**Full Success**:
+- All dimensions reviewed
+- Critical findings = 0
+- High findings ≤ 5
+- Action: Generate final report, mark phase=complete
+
+**Partial Success**:
+- All dimensions reviewed
+- Max iterations reached
+- Still have critical/high findings
+- Action: Generate report with warnings, recommend follow-up
+
+### Error Handling
+
+**Phase-Level Error Matrix**:
+
+| Phase | Error | Blocking? | Action |
+|-------|-------|-----------|--------|
+| Phase 1 | Invalid path pattern | Yes | Error and exit |
+| Phase 1 | No files matched | Yes | Error and exit |
+| Phase 1 | Files not readable | Yes | Error and exit |
+| Phase 2 | Single dimension fails | No | Log warning, continue other dimensions |
+| Phase 2 | All dimensions fail | Yes | Error and exit |
+| Phase 3 | Missing dimension JSON | No | Skip in aggregation, log warning |
+| Phase 4 | Deep-dive agent fails | No | Skip finding, continue others |
+| Phase 4 | Max iterations reached | No | Generate partial report |
+
+**CLI Fallback Chain**: Gemini → Qwen → Codex → degraded mode
+
+**Fallback Triggers**:
+1. HTTP 429, 5xx errors, connection timeout
+2. Invalid JSON output (parse error, missing required fields)
+3. Low confidence score < 0.4
+4. Analysis too brief (< 100 words in report)
+
+**Fallback Behavior**:
+- On trigger: Retry with next tool in chain
+- After Codex fails: Enter degraded mode (skip analysis, log error)
+- Degraded mode: Continue workflow with available results
+
+### TodoWrite Structure
+
+```javascript
+TodoWrite({
+  todos: [
+    { content: "Phase 1: Discovery & Initialization", status: "completed", activeForm: "Initializing" },
+    { content: "Phase 2: Parallel Reviews (7 dimensions)", status: "in_progress", activeForm: "Reviewing" },
+    { content: "  → Security review", status: "in_progress", activeForm: "Analyzing security" },
+    // ... other dimensions as sub-items
+    { content: "Phase 3: Aggregation", status: "pending", activeForm: "Aggregating" },
+    { content: "Phase 4: Deep-dive", status: "pending", activeForm: "Deep-diving" },
+    { content: "Phase 5: Completion", status: "pending", activeForm: "Completing" }
+  ]
+});
+```
+
+## Best Practices
+
+1. **Start Specific**: Begin with focused module patterns for faster results
+2. **Expand Gradually**: Add more modules based on initial findings
+3. **Use Glob Wisely**: `src/auth/**` is more efficient than `src/**` with lots of irrelevant files
+4. **Trust Aggregation Logic**: Auto-selection based on proven heuristics
+5. **Monitor Logs**: Check reports/ directory for CLI analysis insights
+
+## Related Commands
+
+### View Review Progress
+Use `ccw view` to open the review dashboard in browser:
+
+```bash
+ccw view
+```
+
+### Automated Fix Workflow
+After completing a module review, use the generated findings JSON for automated fixing:
+
+```bash
+# Step 1: Complete review (this command)
+/workflow:review-module-cycle src/auth/**
+
+# Step 2: Run automated fixes using dimension findings
+/workflow:review-fix .workflow/active/WFS-{session-id}/.review/
+```
+
+See `/workflow:review-fix` for automated fixing with smart grouping, parallel execution, and test verification.
+

.claude/commands/workflow/review-session-cycle.md

@@ -0,0 +1,782 @@
+---
+name: review-session-cycle
+description: Session-based comprehensive multi-dimensional code review. Analyzes git changes from workflow session across 7 dimensions with hybrid parallel-iterative execution, aggregates findings, and performs focused deep-dives on critical issues until quality gates met.
+argument-hint: "[session-id] [--dimensions=security,architecture,...] [--max-iterations=N]"
+allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*), Task(*)
+---
+
+# Workflow Review-Session-Cycle Command
+
+## Quick Start
+
+```bash
+# Execute comprehensive session review (all 7 dimensions)
+/workflow:review-session-cycle
+
+# Review specific session with custom dimensions
+/workflow:review-session-cycle WFS-payment-integration --dimensions=security,architecture,quality
+
+# Specify session and iteration limit
+/workflow:review-session-cycle WFS-payment-integration --max-iterations=5
+```
+
+**Review Scope**: Git changes from session creation to present (via `git log --since`)
+**Session Requirement**: Requires active or completed workflow session
+**Output Directory**: `.workflow/active/WFS-{session-id}/.review/` (session-based)
+**Default Dimensions**: Security, Architecture, Quality, Action-Items, Performance, Maintainability, Best-Practices
+**Max Iterations**: 3 (adjustable via --max-iterations)
+**Default Iterations**: 1 (deep-dive runs once; use --max-iterations=0 to skip)
+**CLI Tools**: Gemini → Qwen → Codex (fallback chain)
+
+## What & Why
+
+### Core Concept
+Session-based multi-dimensional code review orchestrator with **hybrid parallel-iterative execution** for comprehensive quality assessment of **git changes within a workflow session**.
+
+**Review Scope**:
+- **Session-based**: Reviews only files changed during the workflow session (via `git log --since="${sessionCreatedAt}"`)
+- **For independent module review**: Use `/workflow:review-module-cycle` command instead
+
+**vs Standard Review**:
+- **Standard**: Sequential manual reviews → Inconsistent coverage → Missed cross-cutting concerns
+- **Review-Session-Cycle**: **Parallel automated analysis → Aggregate findings → Deep-dive critical issues** → Comprehensive coverage
+
+### Value Proposition
+1. **Comprehensive Coverage**: 7 specialized dimensions analyze all quality aspects simultaneously
+2. **Intelligent Prioritization**: Automatic identification of critical issues and cross-cutting concerns
+3. **Actionable Insights**: Deep-dive iterations provide step-by-step remediation plans
+
+### Orchestrator Boundary (CRITICAL)
+- **ONLY command** for comprehensive multi-dimensional review
+- Manages: dimension coordination, aggregation, iteration control, progress tracking
+- Delegates: Code exploration and analysis to @cli-explore-agent, dimension-specific reviews via Deep Scan mode
+
+## How It Works
+
+### Execution Flow (Simplified)
+
+```
+Phase 1: Discovery & Initialization
+   └─ Validate session, initialize state, create output structure
+
+Phase 2: Parallel Reviews (for each dimension)
+   ├─ Launch 7 review agents simultaneously
+   ├─ Each executes CLI analysis via Gemini/Qwen
+   ├─ Generate dimension JSON + markdown reports
+   └─ Update review-progress.json
+
+Phase 3: Aggregation
+   ├─ Load all dimension JSON files
+   ├─ Calculate severity distribution (critical/high/medium/low)
+   ├─ Identify cross-cutting concerns (files in 3+ dimensions)
+   └─ Decision:
+      ├─ Critical findings OR high > 5 OR critical files → Phase 4 (Iterate)
+      └─ Else → Phase 5 (Complete)
+
+Phase 4: Iterative Deep-Dive (optional)
+   ├─ Select critical findings (max 5 per iteration)
+   ├─ Launch deep-dive agents for root cause analysis
+   ├─ Generate remediation plans with impact assessment
+   ├─ Re-assess severity based on analysis
+   └─ Loop until no critical findings OR max iterations
+
+Phase 5: Completion
+   └─ Finalize review-progress.json
+```
+
+### Agent Roles
+
+| Agent | Responsibility |
+|-------|---------------|
+| **Orchestrator** | Phase control, session discovery, state management, aggregation logic, iteration control |
+| **@cli-explore-agent** (Review) | Execute dimension-specific code analysis via Deep Scan mode, generate findings JSON with dual-source strategy (Bash + Gemini), create structured analysis reports |
+| **@cli-explore-agent** (Deep-dive) | Focused root cause analysis using dependency mapping, remediation planning with architectural insights, impact assessment, severity re-assessment |
+
+## Enhanced Features
+
+### 1. Review Dimensions Configuration
+
+**7 Specialized Dimensions** with priority-based allocation:
+
+| Dimension | Template | Priority | Timeout |
+|-----------|----------|----------|---------|
+| **Security** | 03-assess-security-risks.txt | 1 (Critical) | 60min |
+| **Architecture** | 02-review-architecture.txt | 2 (High) | 60min |
+| **Quality** | 02-review-code-quality.txt | 3 (Medium) | 40min |
+| **Action-Items** | 02-analyze-code-patterns.txt | 2 (High) | 40min |
+| **Performance** | 03-analyze-performance.txt | 3 (Medium) | 60min |
+| **Maintainability** | 02-review-code-quality.txt* | 3 (Medium) | 40min |
+| **Best-Practices** | 03-review-quality-standards.txt | 3 (Medium) | 40min |
+
+*Custom focus: "Assess technical debt and maintainability"
+
+**Category Definitions by Dimension**:
+
+```javascript
+const CATEGORIES = {
+  security: ['injection', 'authentication', 'authorization', 'encryption', 'input-validation', 'access-control', 'data-exposure'],
+  architecture: ['coupling', 'cohesion', 'layering', 'dependency', 'pattern-violation', 'scalability', 'separation-of-concerns'],
+  quality: ['code-smell', 'duplication', 'complexity', 'naming', 'error-handling', 'testability', 'readability'],
+  'action-items': ['requirement-coverage', 'acceptance-criteria', 'documentation', 'deployment-readiness', 'missing-functionality'],
+  performance: ['n-plus-one', 'inefficient-query', 'memory-leak', 'blocking-operation', 'caching', 'resource-usage'],
+  maintainability: ['technical-debt', 'magic-number', 'long-method', 'large-class', 'dead-code', 'commented-code'],
+  'best-practices': ['convention-violation', 'anti-pattern', 'deprecated-api', 'missing-validation', 'inconsistent-style']
+};
+```
+
+### 2. Aggregation Logic
+
+**Cross-Cutting Concern Detection**:
+1. Files appearing in 3+ dimensions = **Critical Files**
+2. Same issue pattern across dimensions = **Systemic Issue**
+3. Severity clustering in specific files = **Hotspots**
+
+**Deep-Dive Selection Criteria**:
+- All critical severity findings (priority 1)
+- Top 3 high-severity findings in critical files (priority 2)
+- Max 5 findings per iteration (prevent overwhelm)
+
+### 3. Severity Assessment
+
+**Severity Levels**:
+- **Critical**: Security vulnerabilities, data corruption risks, system-wide failures, authentication/authorization bypass
+- **High**: Feature degradation, performance bottlenecks, architecture violations, significant technical debt
+- **Medium**: Code smells, minor performance issues, style inconsistencies, maintainability concerns
+- **Low**: Documentation gaps, minor refactoring opportunities, cosmetic issues
+
+**Iteration Trigger**:
+- Critical findings > 0 OR
+- High findings > 5 OR
+- Critical files count > 0
+
+## Core Responsibilities
+
+### Orchestrator
+
+**Phase 1: Discovery & Initialization**
+
+**Step 1: Session Discovery**
+```javascript
+// If session ID not provided, auto-detect
+if (!providedSessionId) {
+  // Check for active sessions
+  const activeSessions = Glob('.workflow/active/WFS-*');
+  if (activeSessions.length === 1) {
+    sessionId = activeSessions[0].match(/WFS-[^/]+/)[0];
+  } else if (activeSessions.length > 1) {
+    // List sessions and prompt user
+    error("Multiple active sessions found. Please specify session ID.");
+  } else {
+    error("No active session found. Create session first with /workflow:session:start");
+  }
+} else {
+  sessionId = providedSessionId;
+}
+
+// Validate session exists
+Bash(`test -d .workflow/active/${sessionId} && echo "EXISTS"`);
+```
+
+**Step 2: Session Validation**
+- Ensure session has implementation artifacts (check `.summaries/` or `.task/` directory)
+- Extract session creation timestamp from `workflow-session.json`
+- Use timestamp for git log filtering: `git log --since="${sessionCreatedAt}"`
+
+**Step 3: Changed Files Detection**
+```bash
+# Get files changed since session creation
+git log --since="${sessionCreatedAt}" --name-only --pretty=format: | sort -u
+```
+
+**Step 4: Output Directory Setup**
+- Output directory: `.workflow/active/${sessionId}/.review/`
+- Create directory structure:
+  ```bash
+  mkdir -p ${sessionDir}/.review/{dimensions,iterations,reports}
+  ```
+
+**Step 5: Initialize Review State**
+- State initialization: Create `review-state.json` with metadata, dimensions, max_iterations (merged metadata + state)
+- Progress tracking: Create `review-progress.json` for progress tracking
+
+**Step 6: TodoWrite Initialization**
+- Set up progress tracking with hierarchical structure
+- Mark Phase 1 completed, Phase 2 in_progress
+
+**Phase 2: Parallel Review Coordination**
+- Launch 7 @cli-explore-agent instances simultaneously (Deep Scan mode)
+- Pass dimension-specific context (template, timeout, custom focus)
+- Monitor completion via review-progress.json updates
+- TodoWrite updates: Mark dimensions as completed
+- CLI tool fallback: Gemini → Qwen → Codex (on error/timeout)
+
+**Phase 3: Aggregation**
+- Load all dimension JSON files from dimensions/
+- Calculate severity distribution: Count by critical/high/medium/low
+- Identify cross-cutting concerns: Files in 3+ dimensions
+- Select deep-dive findings: Critical + high in critical files (max 5)
+- Decision logic: Iterate if critical > 0 OR high > 5 OR critical files exist
+- Update review-state.json with aggregation results
+
+**Phase 4: Iteration Control**
+- Check iteration count < max_iterations (default 3)
+- Launch deep-dive agents for selected findings
+- Collect remediation plans and re-assessed severities
+- Update severity distribution based on re-assessments
+- Record iteration in review-state.json
+- Loop back to aggregation if still have critical/high findings
+
+**Phase 5: Completion**
+- Finalize review-progress.json with completion statistics
+- Update review-state.json with completion_time and phase=complete
+- TodoWrite completion: Mark all tasks done
+
+
+
+### Session File Structure
+
+```
+.workflow/active/WFS-{session-id}/.review/
+├── review-state.json                    # Orchestrator state machine (includes metadata)
+├── review-progress.json                 # Real-time progress for dashboard
+├── dimensions/                          # Per-dimension results
+│   ├── security.json
+│   ├── architecture.json
+│   ├── quality.json
+│   ├── action-items.json
+│   ├── performance.json
+│   ├── maintainability.json
+│   └── best-practices.json
+├── iterations/                          # Deep-dive results
+│   ├── iteration-1-finding-{uuid}.json
+│   └── iteration-2-finding-{uuid}.json
+└── reports/                             # Human-readable reports
+    ├── security-analysis.md
+    ├── security-cli-output.txt
+    ├── deep-dive-1-{uuid}.md
+    └── ...
+```
+
+**Session Context**:
+```
+.workflow/active/WFS-{session-id}/
+├── workflow-session.json
+├── IMPL_PLAN.md
+├── TODO_LIST.md
+├── .task/
+├── .summaries/
+└── .review/                             # Review results (this command)
+    └── (structure above)
+```
+
+### Review State JSON
+
+**Purpose**: Unified state machine and metadata (merged from metadata + state)
+
+```json
+{
+  "session_id": "WFS-payment-integration",
+  "review_id": "review-20250125-143022",
+  "review_type": "session",
+  "metadata": {
+    "created_at": "2025-01-25T14:30:22Z",
+    "git_changes": {
+      "commit_range": "abc123..def456",
+      "files_changed": 15,
+      "insertions": 342,
+      "deletions": 128
+    },
+    "dimensions": ["security", "architecture", "quality", "action-items", "performance", "maintainability", "best-practices"],
+    "max_iterations": 3
+  },
+  "phase": "parallel|aggregate|iterate|complete",
+  "current_iteration": 1,
+  "dimensions_reviewed": ["security", "architecture", "quality", "action-items", "performance", "maintainability", "best-practices"],
+  "selected_strategy": "comprehensive",
+  "next_action": "execute_parallel_reviews|aggregate_findings|execute_deep_dive|generate_final_report|complete",
+  "severity_distribution": {
+    "critical": 2,
+    "high": 5,
+    "medium": 12,
+    "low": 8
+  },
+  "critical_files": [
+    {
+      "file": "src/payment/processor.ts",
+      "finding_count": 5,
+      "dimensions": ["security", "architecture", "quality"]
+    }
+  ],
+  "iterations": [
+    {
+      "iteration": 1,
+      "findings_analyzed": ["uuid-1", "uuid-2"],
+      "findings_resolved": 1,
+      "findings_escalated": 1,
+      "severity_change": {
+        "before": {"critical": 2, "high": 5, "medium": 12, "low": 8},
+        "after": {"critical": 1, "high": 6, "medium": 12, "low": 8}
+      },
+      "timestamp": "2025-01-25T14:30:00Z"
+    }
+  ],
+  "completion_criteria": {
+    "target": "no_critical_findings_and_high_under_5",
+    "current_status": "in_progress",
+    "estimated_completion": "2 iterations remaining"
+  }
+}
+```
+
+**Field Descriptions**:
+- `phase`: Current execution phase (state machine pointer)
+- `current_iteration`: Iteration counter (used for max check)
+- `next_action`: Next step orchestrator should execute
+- `severity_distribution`: Aggregated counts across all dimensions
+- `critical_files`: Files appearing in 3+ dimensions with metadata
+- `iterations[]`: Historical log for trend analysis
+
+### Review Progress JSON
+
+**Purpose**: Real-time dashboard updates via polling
+
+```json
+{
+  "review_id": "review-20250125-143022",
+  "last_update": "2025-01-25T14:35:10Z",
+  "phase": "parallel|aggregate|iterate|complete",
+  "current_iteration": 1,
+  "progress": {
+    "parallel_review": {
+      "total_dimensions": 7,
+      "completed": 5,
+      "in_progress": 2,
+      "percent_complete": 71
+    },
+    "deep_dive": {
+      "total_findings": 6,
+      "analyzed": 2,
+      "in_progress": 1,
+      "percent_complete": 33
+    }
+  },
+  "agent_status": [
+    {
+      "agent_type": "review-agent",
+      "dimension": "security",
+      "status": "completed",
+      "started_at": "2025-01-25T14:30:00Z",
+      "completed_at": "2025-01-25T15:15:00Z",
+      "duration_ms": 2700000
+    },
+    {
+      "agent_type": "deep-dive-agent",
+      "finding_id": "sec-001-uuid",
+      "status": "in_progress",
+      "started_at": "2025-01-25T14:32:00Z"
+    }
+  ],
+  "estimated_completion": "2025-01-25T16:00:00Z"
+}
+```
+
+### Agent Output Schemas
+
+**Agent-produced JSON files follow standardized schemas**:
+
+1. **Dimension Results** (cli-explore-agent output from parallel reviews)
+   - Schema: `~/.claude/workflows/cli-templates/schemas/review-dimension-results-schema.json`
+   - Output: `.review-cycle/dimensions/{dimension}.json`
+   - Contains: findings array, summary statistics, cross_references
+
+2. **Deep-Dive Results** (cli-explore-agent output from iterations)
+   - Schema: `~/.claude/workflows/cli-templates/schemas/review-deep-dive-results-schema.json`
+   - Output: `.review-cycle/iterations/iteration-{N}-finding-{uuid}.json`
+   - Contains: root_cause, remediation_plan, impact_assessment, reassessed_severity
+
+### Agent Invocation Template
+
+**Review Agent** (parallel execution, 7 instances):
+
+```javascript
+Task(
+  subagent_type="cli-explore-agent",
+  run_in_background=false,
+  description=`Execute ${dimension} review analysis via Deep Scan`,
+  prompt=`
+    ## Task Objective
+    Conduct comprehensive ${dimension} code exploration and analysis using Deep Scan mode (Bash + Gemini dual-source strategy) for completed implementation in session ${sessionId}
+
+    ## Analysis Mode Selection
+    Use **Deep Scan mode** for this review:
+    - Phase 1: Bash structural scan for standard patterns (classes, functions, imports)
+    - Phase 2: Gemini semantic analysis for design intent, non-standard patterns, ${dimension}-specific concerns
+    - Phase 3: Synthesis with attribution (bash-discovered vs gemini-discovered findings)
+
+    ## MANDATORY FIRST STEPS (Execute by Agent)
+    **You (cli-explore-agent) MUST execute these steps in order:**
+    1. Read session metadata: ${sessionMetadataPath}
+    2. Read completed task summaries: bash(find ${summariesDir} -name "IMPL-*.md" -type f)
+    3. Get changed files: bash(cd ${workflowDir} && git log --since="${sessionCreatedAt}" --name-only --pretty=format: | sort -u)
+    4. Read review state: ${reviewStateJsonPath}
+    5. Execute: cat ~/.claude/workflows/cli-templates/schemas/review-dimension-results-schema.json (get output schema reference)
+    6. Read: .workflow/project-tech.json (technology stack and architecture context)
+    7. Read: .workflow/project-guidelines.json (user-defined constraints and conventions to validate against)
+
+    ## Session Context
+    - Session ID: ${sessionId}
+    - Review Dimension: ${dimension}
+    - Review ID: ${reviewId}
+    - Implementation Phase: Complete (all tests passing)
+    - Output Directory: ${outputDir}
+
+    ## CLI Configuration
+    - Tool Priority: gemini → qwen → codex (fallback chain)
+    - Template: ~/.claude/workflows/cli-templates/prompts/analysis/${dimensionTemplate}
+    - Custom Focus: ${customFocus || 'Standard dimension analysis'}
+    - Timeout: ${timeout}ms
+    - Mode: analysis (READ-ONLY)
+
+    ## Expected Deliverables
+
+    **Schema Reference**: Schema obtained in MANDATORY FIRST STEPS step 5, follow schema exactly
+
+    1. Dimension Results JSON: ${outputDir}/dimensions/${dimension}.json
+
+       **⚠️ CRITICAL JSON STRUCTURE REQUIREMENTS**:
+
+       Root structure MUST be array: \`[{ ... }]\` NOT \`{ ... }\`
+
+       Required top-level fields:
+       - dimension, review_id, analysis_timestamp (NOT timestamp/analyzed_at)
+       - cli_tool_used (gemini|qwen|codex), model, analysis_duration_ms
+       - summary (FLAT structure), findings, cross_references
+
+       Summary MUST be FLAT (NOT nested by_severity):
+       \`{ "total_findings": N, "critical": N, "high": N, "medium": N, "low": N, "files_analyzed": N, "lines_reviewed": N }\`
+
+       Finding required fields:
+       - id: format \`{dim}-{seq}-{uuid8}\` e.g., \`sec-001-a1b2c3d4\` (lowercase)
+       - severity: lowercase only (critical|high|medium|low)
+       - snippet (NOT code_snippet), impact (NOT exploit_scenario)
+       - metadata, iteration (0), status (pending_remediation), cross_references
+
+    2. Analysis Report: ${outputDir}/reports/${dimension}-analysis.md
+       - Human-readable summary with recommendations
+       - Grouped by severity: critical → high → medium → low
+       - Include file:line references for all findings
+
+    3. CLI Output Log: ${outputDir}/reports/${dimension}-cli-output.txt
+       - Raw CLI tool output for debugging
+       - Include full analysis text
+
+    ## Dimension-Specific Guidance
+    ${getDimensionGuidance(dimension)}
+
+    ## Success Criteria
+    - [ ] Schema obtained via cat review-dimension-results-schema.json
+    - [ ] All changed files analyzed for ${dimension} concerns
+    - [ ] All findings include file:line references with code snippets
+    - [ ] Severity assessment follows established criteria (see reference)
+    - [ ] Recommendations are actionable with code examples
+    - [ ] JSON output follows schema exactly
+    - [ ] Report is comprehensive and well-organized
+  `
+)
+```
+
+**Deep-Dive Agent** (iteration execution):
+
+```javascript
+Task(
+  subagent_type="cli-explore-agent",
+  run_in_background=false,
+  description=`Deep-dive analysis for critical finding: ${findingTitle} via Dependency Map + Deep Scan`,
+  prompt=`
+    ## Task Objective
+    Perform focused root cause analysis using Dependency Map mode (for impact analysis) + Deep Scan mode (for semantic understanding) to generate comprehensive remediation plan for critical ${dimension} issue
+
+    ## Analysis Mode Selection
+    Use **Dependency Map mode** first to understand dependencies:
+    - Build dependency graph around ${file} to identify affected components
+    - Detect circular dependencies or tight coupling related to this finding
+    - Calculate change risk scores for remediation impact
+
+    Then apply **Deep Scan mode** for semantic analysis:
+    - Understand design intent and architectural context
+    - Identify non-standard patterns or implicit dependencies
+    - Extract remediation insights from code structure
+
+    ## Finding Context
+    - Finding ID: ${findingId}
+    - Original Dimension: ${dimension}
+    - Title: ${findingTitle}
+    - File: ${file}:${line}
+    - Severity: ${severity}
+    - Category: ${category}
+    - Original Description: ${description}
+    - Iteration: ${iteration}
+
+    ## MANDATORY FIRST STEPS (Execute by Agent)
+    **You (cli-explore-agent) MUST execute these steps in order:**
+    1. Read original finding: ${dimensionJsonPath}
+    2. Read affected file: ${file}
+    3. Identify related code: bash(grep -r "import.*${basename(file)}" ${workflowDir}/src --include="*.ts")
+    4. Read test files: bash(find ${workflowDir}/tests -name "*${basename(file, '.ts')}*" -type f)
+    5. Execute: cat ~/.claude/workflows/cli-templates/schemas/review-deep-dive-results-schema.json (get output schema reference)
+    6. Read: .workflow/project-tech.json (technology stack and architecture context)
+    7. Read: .workflow/project-guidelines.json (user-defined constraints for remediation compliance)
+
+    ## CLI Configuration
+    - Tool Priority: gemini → qwen → codex
+    - Template: ~/.claude/workflows/cli-templates/prompts/analysis/01-diagnose-bug-root-cause.txt
+    - Timeout: 2400000ms (40 minutes)
+    - Mode: analysis (READ-ONLY)
+
+    ## Expected Deliverables
+
+    **Schema Reference**: Schema obtained in MANDATORY FIRST STEPS step 5, follow schema exactly
+
+    1. Deep-Dive Results JSON: ${outputDir}/iterations/iteration-${iteration}-finding-${findingId}.json
+
+       **⚠️ CRITICAL JSON STRUCTURE REQUIREMENTS**:
+
+       Root structure MUST be array: \`[{ ... }]\` NOT \`{ ... }\`
+
+       Required top-level fields:
+       - finding_id, dimension, iteration, analysis_timestamp
+       - cli_tool_used, model, analysis_duration_ms
+       - original_finding, root_cause, remediation_plan
+       - impact_assessment, reassessed_severity, confidence_score, cross_references
+
+       All nested objects must follow schema exactly - read schema for field names
+
+    2. Analysis Report: ${outputDir}/reports/deep-dive-${iteration}-${findingId}.md
+       - Detailed root cause analysis
+       - Step-by-step remediation plan
+       - Impact assessment and rollback strategy
+
+    ## Success Criteria
+    - [ ] Schema obtained via cat review-deep-dive-results-schema.json
+    - [ ] Root cause clearly identified with supporting evidence
+    - [ ] Remediation plan is step-by-step actionable with exact file:line references
+    - [ ] Each step includes specific commands and validation tests
+    - [ ] Impact fully assessed (files, tests, breaking changes, dependencies)
+    - [ ] Severity re-evaluation justified with evidence
+    - [ ] Confidence score accurately reflects certainty of analysis
+    - [ ] JSON output follows schema exactly
+    - [ ] References include project-specific and external documentation
+  `
+)
+```
+
+### Dimension Guidance Reference
+
+```javascript
+function getDimensionGuidance(dimension) {
+  const guidance = {
+    security: `
+      Focus Areas:
+      - Input validation and sanitization
+      - Authentication and authorization mechanisms
+      - Data encryption (at-rest and in-transit)
+      - SQL/NoSQL injection vulnerabilities
+      - XSS, CSRF, and other web vulnerabilities
+      - Sensitive data exposure
+      - Access control and privilege escalation
+
+      Severity Criteria:
+      - Critical: Authentication bypass, SQL injection, RCE, sensitive data exposure
+      - High: Missing authorization checks, weak encryption, exposed secrets
+      - Medium: Missing input validation, insecure defaults, weak password policies
+      - Low: Security headers missing, verbose error messages, outdated dependencies
+    `,
+    architecture: `
+      Focus Areas:
+      - Layering and separation of concerns
+      - Coupling and cohesion
+      - Design pattern adherence
+      - Dependency management
+      - Scalability and extensibility
+      - Module boundaries
+      - API design consistency
+
+      Severity Criteria:
+      - Critical: Circular dependencies, god objects, tight coupling across layers
+      - High: Violated architectural principles, scalability bottlenecks
+      - Medium: Missing abstractions, inconsistent patterns, suboptimal design
+      - Low: Minor coupling issues, documentation gaps, naming inconsistencies
+    `,
+    quality: `
+      Focus Areas:
+      - Code duplication
+      - Complexity (cyclomatic, cognitive)
+      - Naming conventions
+      - Error handling patterns
+      - Code readability
+      - Comment quality
+      - Dead code
+
+      Severity Criteria:
+      - Critical: Severe complexity (CC > 20), massive duplication (>50 lines)
+      - High: High complexity (CC > 10), significant duplication, poor error handling
+      - Medium: Moderate complexity (CC > 5), naming issues, code smells
+      - Low: Minor duplication, documentation gaps, cosmetic issues
+    `,
+    'action-items': `
+      Focus Areas:
+      - Requirements coverage verification
+      - Acceptance criteria met
+      - Documentation completeness
+      - Deployment readiness
+      - Missing functionality
+      - Test coverage gaps
+      - Configuration management
+
+      Severity Criteria:
+      - Critical: Core requirements not met, deployment blockers
+      - High: Significant functionality missing, acceptance criteria not met
+      - Medium: Minor requirements gaps, documentation incomplete
+      - Low: Nice-to-have features missing, minor documentation gaps
+    `,
+    performance: `
+      Focus Areas:
+      - N+1 query problems
+      - Inefficient algorithms (O(n²) where O(n log n) possible)
+      - Memory leaks
+      - Blocking operations on main thread
+      - Missing caching opportunities
+      - Resource usage (CPU, memory, network)
+      - Database query optimization
+
+      Severity Criteria:
+      - Critical: Memory leaks, O(n²) in hot path, blocking main thread
+      - High: N+1 queries, missing indexes, inefficient algorithms
+      - Medium: Suboptimal caching, unnecessary computations, lazy loading issues
+      - Low: Minor optimization opportunities, redundant operations
+    `,
+    maintainability: `
+      Focus Areas:
+      - Technical debt indicators
+      - Magic numbers and hardcoded values
+      - Long methods (>50 lines)
+      - Large classes (>500 lines)
+      - Dead code and commented code
+      - Code documentation
+      - Test coverage
+
+      Severity Criteria:
+      - Critical: Massive methods (>200 lines), severe technical debt blocking changes
+      - High: Large methods (>100 lines), significant dead code, undocumented complex logic
+      - Medium: Magic numbers, moderate technical debt, missing tests
+      - Low: Minor refactoring opportunities, cosmetic improvements
+    `,
+    'best-practices': `
+      Focus Areas:
+      - Framework conventions adherence
+      - Language idioms
+      - Anti-patterns
+      - Deprecated API usage
+      - Coding standards compliance
+      - Error handling patterns
+      - Logging and monitoring
+
+      Severity Criteria:
+      - Critical: Severe anti-patterns, deprecated APIs with security risks
+      - High: Major convention violations, poor error handling, missing logging
+      - Medium: Minor anti-patterns, style inconsistencies, suboptimal patterns
+      - Low: Cosmetic style issues, minor convention deviations
+    `
+  };
+
+  return guidance[dimension] || 'Standard code review analysis';
+}
+```
+
+### Completion Conditions
+
+**Full Success**:
+- All dimensions reviewed
+- Critical findings = 0
+- High findings ≤ 5
+- Action: Generate final report, mark phase=complete
+
+**Partial Success**:
+- All dimensions reviewed
+- Max iterations reached
+- Still have critical/high findings
+- Action: Generate report with warnings, recommend follow-up
+
+### Error Handling
+
+**Phase-Level Error Matrix**:
+
+| Phase | Error | Blocking? | Action |
+|-------|-------|-----------|--------|
+| Phase 1 | Session not found | Yes | Error and exit |
+| Phase 1 | No completed tasks | Yes | Error and exit |
+| Phase 1 | No changed files | Yes | Error and exit |
+| Phase 2 | Single dimension fails | No | Log warning, continue other dimensions |
+| Phase 2 | All dimensions fail | Yes | Error and exit |
+| Phase 3 | Missing dimension JSON | No | Skip in aggregation, log warning |
+| Phase 4 | Deep-dive agent fails | No | Skip finding, continue others |
+| Phase 4 | Max iterations reached | No | Generate partial report |
+
+**CLI Fallback Chain**: Gemini → Qwen → Codex → degraded mode
+
+**Fallback Triggers**:
+1. HTTP 429, 5xx errors, connection timeout
+2. Invalid JSON output (parse error, missing required fields)
+3. Low confidence score < 0.4
+4. Analysis too brief (< 100 words in report)
+
+**Fallback Behavior**:
+- On trigger: Retry with next tool in chain
+- After Codex fails: Enter degraded mode (skip analysis, log error)
+- Degraded mode: Continue workflow with available results
+
+### TodoWrite Structure
+
+```javascript
+TodoWrite({
+  todos: [
+    { content: "Phase 1: Discovery & Initialization", status: "completed", activeForm: "Initializing" },
+    { content: "Phase 2: Parallel Reviews (7 dimensions)", status: "in_progress", activeForm: "Reviewing" },
+    { content: "  → Security review", status: "in_progress", activeForm: "Analyzing security" },
+    // ... other dimensions as sub-items
+    { content: "Phase 3: Aggregation", status: "pending", activeForm: "Aggregating" },
+    { content: "Phase 4: Deep-dive", status: "pending", activeForm: "Deep-diving" },
+    { content: "Phase 5: Completion", status: "pending", activeForm: "Completing" }
+  ]
+});
+```
+
+## Best Practices
+
+1. **Default Settings Work**: 7 dimensions + 3 iterations sufficient for most cases
+2. **Parallel Execution**: ~60 minutes for full initial review (7 dimensions)
+3. **Trust Aggregation Logic**: Auto-selection based on proven heuristics
+4. **Monitor Logs**: Check reports/ directory for CLI analysis insights
+
+## Related Commands
+
+### View Review Progress
+Use `ccw view` to open the review dashboard in browser:
+
+```bash
+ccw view
+```
+
+### Automated Fix Workflow
+After completing a review, use the generated findings JSON for automated fixing:
+
+```bash
+# Step 1: Complete review (this command)
+/workflow:review-session-cycle
+
+# Step 2: Run automated fixes using dimension findings
+/workflow:review-fix .workflow/active/WFS-{session-id}/.review/
+```
+
+See `/workflow:review-fix` for automated fixing with smart grouping, parallel execution, and test verification.
+

.claude/commands/workflow/review.md

@@ -1,6 +1,6 @@
 ---
 name: review
-description: Optional specialized review (security, architecture, docs) for completed implementation
+description: Post-implementation review with specialized types (security/architecture/action-items/quality) using analysis agents and Gemini
 argument-hint: "[--type=security|architecture|action-items|quality] [optional: session-id]"
 ---
 
@@ -29,6 +29,39 @@ argument-hint: "[--type=security|architecture|action-items|quality] [optional: s
 - For documentation generation, use `/workflow:tools:docs`
 - For CLAUDE.md updates, use `/update-memory-related`
 
+## Execution Process
+
+```
+Input Parsing:
+   ├─ Parse --type flag (default: quality)
+   └─ Parse session-id argument (optional)
+
+Step 1: Session Resolution
+   └─ Decision:
+      ├─ session-id provided → Use provided session
+      └─ Not provided → Auto-detect from .workflow/active/
+
+Step 2: Validation
+   ├─ Check session directory exists
+   └─ Check for completed implementation (.summaries/IMPL-*.md exists)
+
+Step 3: Type Check
+   └─ Decision:
+      ├─ type=docs → Redirect to /workflow:tools:docs
+      └─ Other types → Continue to analysis
+
+Step 4: Model Analysis Phase
+   ├─ Load context (summaries, test results, changed files)
+   └─ Perform specialized review by type:
+      ├─ security → Security patterns + Gemini analysis
+      ├─ architecture → Qwen architecture analysis
+      ├─ quality → Gemini code quality analysis
+      └─ action-items → Requirements verification
+
+Step 5: Generate Report
+   └─ Output: REVIEW-{type}.md
+```
+
 ## Execution Template
 
 ```bash
@@ -39,17 +72,17 @@ argument-hint: "[--type=security|architecture|action-items|quality] [optional: s
 if [ -n "$SESSION_ARG" ]; then
     sessionId="$SESSION_ARG"
 else
-    sessionId=$(find .workflow/ -name '.active-*' | head -1 | sed 's/.*active-//')
+    sessionId=$(find .workflow/active/ -name "WFS-*" -type d | head -1 | xargs basename)
 fi
 
 # Step 2: Validation
-if [ ! -d ".workflow/${sessionId}" ]; then
+if [ ! -d ".workflow/active/${sessionId}" ]; then
     echo "Session ${sessionId} not found"
     exit 1
 fi
 
 # Check for completed tasks
-if [ ! -d ".workflow/${sessionId}/.summaries" ] || [ -z "$(find .workflow/${sessionId}/.summaries/ -name "IMPL-*.md" -type f 2>/dev/null)" ]; then
+if [ ! -d ".workflow/active/${sessionId}/.summaries" ] || [ -z "$(find .workflow/active/${sessionId}/.summaries/ -name "IMPL-*.md" -type f 2>/dev/null)" ]; then
     echo "No completed implementation found. Complete implementation first"
     exit 1
 fi
@@ -79,14 +112,18 @@ After bash validation, the model takes control to:
 
 1. **Load Context**: Read completed task summaries and changed files
    ```bash
-   # Load implementation summaries
-   cat .workflow/${sessionId}/.summaries/IMPL-*.md
+   # Load implementation summaries (iterate through .summaries/ directory)
+   for summary in .workflow/active/${sessionId}/.summaries/*.md; do
+     cat "$summary"
+   done
 
    # Load test results (if available)
-   cat .workflow/${sessionId}/.summaries/TEST-FIX-*.md 2>/dev/null
+   for test_summary in .workflow/active/${sessionId}/.summaries/TEST-FIX-*.md 2>/dev/null; do
+     cat "$test_summary"
+   done
 
    # Get changed files
-   git log --since="$(cat .workflow/${sessionId}/workflow-session.json | jq -r .created_at)" --name-only --pretty=format: | sort -u
+   git log --since="$(cat .workflow/active/${sessionId}/workflow-session.json | jq -r .created_at)" --name-only --pretty=format: | sort -u
    ```
 
 2. **Perform Specialized Review**: Based on `review_type`
@@ -99,54 +136,56 @@ After bash validation, the model takes control to:
      ```
    - Use Gemini for security analysis:
      ```bash
-     cd .workflow/${sessionId} && gemini -p "
+     ccw cli -p "
      PURPOSE: Security audit of completed implementation
      TASK: Review code for security vulnerabilities, insecure patterns, auth/authz issues
-     CONTEXT: @.summaries/IMPL-*.md,../.. @../../CLAUDE.md
+     CONTEXT: @.summaries/IMPL-*.md,../.. @../../project-tech.json @../../project-guidelines.json
      EXPECTED: Security findings report with severity levels
      RULES: Focus on OWASP Top 10, authentication, authorization, data validation, injection risks
-     " --approval-mode yolo
+     " --tool gemini --mode write --cd .workflow/active/${sessionId}
      ```
 
    **Architecture Review** (`--type=architecture`):
    - Use Qwen for architecture analysis:
      ```bash
-     cd .workflow/${sessionId} && qwen -p "
+     ccw cli -p "
      PURPOSE: Architecture compliance review
      TASK: Evaluate adherence to architectural patterns, identify technical debt, review design decisions
-     CONTEXT: @.summaries/IMPL-*.md,../.. @../../CLAUDE.md
+     CONTEXT: @.summaries/IMPL-*.md,../.. @../../project-tech.json @../../project-guidelines.json
      EXPECTED: Architecture assessment with recommendations
      RULES: Check for patterns, separation of concerns, modularity, scalability
-     " --approval-mode yolo
+     " --tool qwen --mode write --cd .workflow/active/${sessionId}
      ```
 
    **Quality Review** (`--type=quality`):
    - Use Gemini for code quality:
      ```bash
-     cd .workflow/${sessionId} && gemini -p "
+     ccw cli -p "
      PURPOSE: Code quality and best practices review
      TASK: Assess code readability, maintainability, adherence to best practices
-     CONTEXT: @.summaries/IMPL-*.md,../.. @../../CLAUDE.md
+     CONTEXT: @.summaries/IMPL-*.md,../.. @../../project-tech.json @../../project-guidelines.json
      EXPECTED: Quality assessment with improvement suggestions
      RULES: Check for code smells, duplication, complexity, naming conventions
-     " --approval-mode yolo
+     " --tool gemini --mode write --cd .workflow/active/${sessionId}
      ```
 
    **Action Items Review** (`--type=action-items`):
    - Verify all requirements and acceptance criteria met:
      ```bash
      # Load task requirements and acceptance criteria
-     find .workflow/${sessionId}/.task -name "IMPL-*.json" -exec jq -r '
-       "Task: " + .id + "\n" +
-       "Requirements: " + (.context.requirements | join(", ")) + "\n" +
-       "Acceptance: " + (.context.acceptance | join(", "))
-     ' {} \;
+     for task_file in .workflow/active/${sessionId}/.task/*.json; do
+       cat "$task_file" | jq -r '
+         "Task: " + .id + "\n" +
+         "Requirements: " + (.context.requirements | join(", ")) + "\n" +
+         "Acceptance: " + (.context.acceptance | join(", "))
+       '
+     done
 
      # Check implementation summaries against requirements
-     cd .workflow/${sessionId} && gemini -p "
+     ccw cli -p "
      PURPOSE: Verify all requirements and acceptance criteria are met
      TASK: Cross-check implementation summaries against original requirements
-     CONTEXT: @.task/IMPL-*.json,.summaries/IMPL-*.md,../.. @../../CLAUDE.md
+     CONTEXT: @.task/IMPL-*.json,.summaries/IMPL-*.md,../.. @../../project-tech.json @../../project-guidelines.json
      EXPECTED:
      - Requirements coverage matrix
      - Acceptance criteria verification
@@ -157,7 +196,7 @@ After bash validation, the model takes control to:
      - Verify all acceptance criteria are met
      - Flag any incomplete or missing action items
      - Assess deployment readiness
-     " --approval-mode yolo
+     " --tool gemini --mode write --cd .workflow/active/${sessionId}
      ```
 
 
@@ -195,7 +234,7 @@ After bash validation, the model takes control to:
 4. **Output Files**:
    ```bash
    # Save review report
-   Write(.workflow/${sessionId}/REVIEW-${review_type}.md)
+   Write(.workflow/active/${sessionId}/REVIEW-${review_type}.md)
 
    # Update session metadata
    # (optional) Update workflow-session.json with review status

.claude/commands/workflow/session/complete.md

@@ -1,6 +1,6 @@
 ---
 name: complete
-description: Mark the active workflow session as complete, archive it with lessons learned, and remove active flag
+description: Mark active workflow session as complete, archive with lessons learned, update manifest, remove active flag
 examples:
   - /workflow:session:complete
   - /workflow:session:complete --detailed
@@ -8,140 +8,146 @@ examples:
 
 # Complete Workflow Session (/workflow:session:complete)
 
-## Overview
-Mark the currently active workflow session as complete, analyze it for lessons learned, move it to the archive directory, and remove the active flag marker.
+Mark the currently active workflow session as complete, archive it, and update manifests.
 
-## Usage
-```bash
-/workflow:session:complete           # Complete current active session
-/workflow:session:complete --detailed # Show detailed completion summary
-```
-
-## Implementation Flow
-
-### Phase 1: Prepare for Archival (Minimal Manual Operations)
-
-**Purpose**: Find active session, move to archive location, pass control to agent. Minimal operations.
-
-#### Step 1.1: Find Active Session and Get Name
-```bash
-# Find active marker
-bash(find .workflow/ -name ".active-*" -type f | head -1)
-
-# Extract session name from marker path
-bash(basename .workflow/.active-WFS-session-name | sed 's/^\.active-//')
-```
-**Output**: Session name `WFS-session-name`
-
-#### Step 1.2: Move Session to Archive
-```bash
-# Create archive directory if needed
-bash(mkdir -p .workflow/.archives/)
-
-# Move session to archive location
-bash(mv .workflow/WFS-session-name .workflow/.archives/WFS-session-name)
-```
-**Result**: Session now at `.workflow/.archives/WFS-session-name/`
-
-### Phase 2: Agent-Orchestrated Completion (All Data Processing)
-
-**Purpose**: Agent analyzes archived session, generates metadata, updates manifest, and removes active marker.
-
-#### Agent Invocation
-
-Invoke `universal-executor` agent to complete the archival process.
-
-**Agent Task**:
-```
-Task(
-  subagent_type="universal-executor",
-  description="Complete session archival",
-  prompt=`
-Complete workflow session archival. Session already moved to archive location.
-
-## Context
-- Session: .workflow/.archives/WFS-session-name/
-- Active marker: .workflow/.active-WFS-session-name
-
-## Tasks
-
-1. **Extract session data** from workflow-session.json (session_id, description/topic, started_at/timestamp, completed_at, status)
-   - If status != "completed", update it with timestamp
-
-2. **Count files**: tasks (.task/*.json) and summaries (.summaries/*.md)
-
-3. **Generate lessons**: Use gemini with ~/.claude/workflows/cli-templates/prompts/archive/analysis-simple.txt (fallback: analyze files directly)
-   - Return: {successes, challenges, watch_patterns}
-
-4. **Build archive entry**:
-   - Calculate: duration_hours, success_rate, tags (3-5 keywords)
-   - Construct complete JSON with session_id, description, archived_at, archive_path, metrics, tags, lessons
-
-5. **Update manifest**: Initialize .workflow/.archives/manifest.json if needed, append entry
-
-6. **Remove active marker**
-
-7. **Return result**: {"status": "success", "session_id": "...", "archived_at": "...", "metrics": {...}, "lessons_summary": {...}}
-
-## Error Handling
-- On failure: return {"status": "error", "task": "...", "message": "..."}
-- Do NOT remove marker if failed
-  `
-)
-```
-
-**Expected Output**:
-- Agent returns JSON result confirming successful archival
-- Display completion summary to user based on agent response
-
-## Workflow Execution Strategy
-
-### Two-Phase Approach (Optimized)
-
-**Phase 1: Minimal Manual Setup** (2 simple operations)
-- Find active session and extract name
-- Move session to archive location
-- **No data extraction** - agent handles all data processing
-- **No counting** - agent does this from archive location
-- **Total**: 2 bash commands (find + move)
-
-**Phase 2: Agent-Driven Completion** (1 agent invocation)
-- Extract all session data from archived location
-- Count tasks and summaries
-- Generate lessons learned analysis
-- Build complete archive metadata
-- Update manifest
-- Remove active marker
-- Return success/error result
-
-## Quick Commands
+## Pre-defined Commands
 
 ```bash
-# Phase 1: Find and move
-bash(find .workflow/ -name ".active-*" -type f | head -1)
-bash(basename .workflow/.active-WFS-session-name | sed 's/^\.active-//')
-bash(mkdir -p .workflow/.archives/)
-bash(mv .workflow/WFS-session-name .workflow/.archives/WFS-session-name)
+# Phase 1: Find active session
+SESSION_PATH=$(find .workflow/active/ -maxdepth 1 -name "WFS-*" -type d | head -1)
+SESSION_ID=$(basename "$SESSION_PATH")
 
-# Phase 2: Agent completes archival
-Task(subagent_type="universal-executor", description="Complete session archival", prompt=`...`)
+# Phase 3: Move to archive
+mkdir -p .workflow/archives/
+mv .workflow/active/$SESSION_ID .workflow/archives/$SESSION_ID
+
+# Cleanup marker
+rm -f .workflow/archives/$SESSION_ID/.archiving
 ```
 
-## Archive Query Commands
+## Key Files to Read
 
-After archival, you can query the manifest:
+**For manifest.json generation**, read ONLY these files:
+
+| File | Extract |
+|------|---------|
+| `$SESSION_PATH/workflow-session.json` | session_id, description, started_at, status |
+| `$SESSION_PATH/IMPL_PLAN.md` | title (first # heading), description (first paragraph) |
+| `$SESSION_PATH/.tasks/*.json` | count files |
+| `$SESSION_PATH/.summaries/*.md` | count files |
+| `$SESSION_PATH/.review/dimensions/*.json` | count + findings summary (optional) |
+
+## Execution Flow
+
+### Phase 1: Find Session (2 commands)
 
 ```bash
-# List all archived sessions
-jq '.archives[].session_id' .workflow/.archives/manifest.json
+# 1. Find and extract session
+SESSION_PATH=$(find .workflow/active/ -maxdepth 1 -name "WFS-*" -type d | head -1)
+SESSION_ID=$(basename "$SESSION_PATH")
 
-# Find sessions by keyword
-jq '.archives[] | select(.description | test("auth"; "i"))' .workflow/.archives/manifest.json
-
-# Get specific session details
-jq '.archives[] | select(.session_id == "WFS-user-auth")' .workflow/.archives/manifest.json
-
-# List all watch patterns across sessions
-jq '.archives[].lessons.watch_patterns[]' .workflow/.archives/manifest.json
+# 2. Check/create archiving marker
+test -f "$SESSION_PATH/.archiving" && echo "RESUMING" || touch "$SESSION_PATH/.archiving"
 ```
 
+**Output**: `SESSION_ID` = e.g., `WFS-auth-feature`
+
+### Phase 2: Generate Manifest Entry (Read-only)
+
+Read the key files above, then build this structure:
+
+```json
+{
+  "session_id": "<from workflow-session.json>",
+  "description": "<from workflow-session.json>",
+  "archived_at": "<current ISO timestamp>",
+  "archive_path": ".workflow/archives/<SESSION_ID>",
+  "metrics": {
+    "duration_hours": "<(completed_at - started_at) / 3600000>",
+    "tasks_completed": "<count .tasks/*.json>",
+    "summaries_generated": "<count .summaries/*.md>",
+    "review_metrics": {
+      "dimensions_analyzed": "<count .review/dimensions/*.json>",
+      "total_findings": "<sum from dimension JSONs>"
+    }
+  },
+  "tags": ["<3-5 keywords from IMPL_PLAN.md>"],
+  "lessons": {
+    "successes": ["<key wins>"],
+    "challenges": ["<difficulties>"],
+    "watch_patterns": ["<patterns to monitor>"]
+  }
+}
+```
+
+**Lessons Generation**: Use gemini with `~/.claude/workflows/cli-templates/prompts/archive/analysis-simple.txt`
+
+### Phase 3: Atomic Commit (4 commands)
+
+```bash
+# 1. Create archive directory
+mkdir -p .workflow/archives/
+
+# 2. Move session
+mv .workflow/active/$SESSION_ID .workflow/archives/$SESSION_ID
+
+# 3. Update manifest.json (Read → Append → Write)
+# Read: .workflow/archives/manifest.json (or [])
+# Append: archive_entry from Phase 2
+# Write: updated JSON
+
+# 4. Remove marker
+rm -f .workflow/archives/$SESSION_ID/.archiving
+```
+
+**Output**:
+```
+✓ Session "$SESSION_ID" archived successfully
+  Location: .workflow/archives/$SESSION_ID/
+  Manifest: Updated with N total sessions
+```
+
+### Phase 4: Update project.json (Optional)
+
+**Skip if**: `.workflow/project.json` doesn't exist
+
+```bash
+# Check
+test -f .workflow/project.json || echo "SKIP"
+```
+
+**If exists**, add feature entry:
+
+```json
+{
+  "id": "<slugified title>",
+  "title": "<from IMPL_PLAN.md>",
+  "status": "completed",
+  "tags": ["<from Phase 2>"],
+  "timeline": { "implemented_at": "<date>" },
+  "traceability": { "session_id": "<SESSION_ID>", "archive_path": "<path>" }
+}
+```
+
+**Output**:
+```
+✓ Feature added to project registry
+```
+
+## Error Recovery
+
+| Phase | Symptom | Recovery |
+|-------|---------|----------|
+| 1 | No active session | `No active session found` |
+| 2 | Analysis fails | Remove marker: `rm $SESSION_PATH/.archiving`, retry |
+| 3 | Move fails | Session safe in active/, fix issue, retry |
+| 3 | Manifest fails | Session in archives/, manually add entry, remove marker |
+
+## Quick Reference
+
+```
+Phase 1: find session → create .archiving marker
+Phase 2: read key files → build manifest entry (no writes)
+Phase 3: mkdir → mv → update manifest.json → rm marker
+Phase 4: update project.json features array (optional)
+```

.claude/commands/workflow/session/list.md

@@ -1,6 +1,6 @@
 ---
 name: list
-description: List all workflow sessions with status
+description: List all workflow sessions with status filtering, shows session metadata and progress information
 examples:
   - /workflow:session:list
 ---
@@ -19,35 +19,35 @@ Display all workflow sessions with their current status, progress, and metadata.
 
 ### Step 1: Find All Sessions
 ```bash
-ls .workflow/WFS-* 2>/dev/null
+ls .workflow/active/WFS-* 2>/dev/null
 ```
 
 ### Step 2: Check Active Session
 ```bash
-ls .workflow/.active-* 2>/dev/null | head -1
+find .workflow/active/ -name "WFS-*" -type d 2>/dev/null | head -1
 ```
 
 ### Step 3: Read Session Metadata
 ```bash
-jq -r '.session_id, .status, .project' .workflow/WFS-session/workflow-session.json
+jq -r '.session_id, .status, .project' .workflow/active/WFS-session/workflow-session.json
 ```
 
 ### Step 4: Count Task Progress
 ```bash
-find .workflow/WFS-session/.task/ -name "*.json" -type f 2>/dev/null | wc -l
-find .workflow/WFS-session/.summaries/ -name "*.md" -type f 2>/dev/null | wc -l
+find .workflow/active/WFS-session/.task/ -name "*.json" -type f 2>/dev/null | wc -l
+find .workflow/active/WFS-session/.summaries/ -name "*.md" -type f 2>/dev/null | wc -l
 ```
 
 ### Step 5: Get Creation Time
 ```bash
-jq -r '.created_at // "unknown"' .workflow/WFS-session/workflow-session.json
+jq -r '.created_at // "unknown"' .workflow/active/WFS-session/workflow-session.json
 ```
 
 ## Simple Bash Commands
 
 ### Basic Operations
-- **List sessions**: `find .workflow/ -maxdepth 1 -type d -name "WFS-*"`
-- **Find active**: `find .workflow/ -name ".active-*" -type f`
+- **List sessions**: `find .workflow/active/ -name "WFS-*" -type d`
+- **Find active**: `find .workflow/active/ -name "WFS-*" -type d`
 - **Read session data**: `jq -r '.session_id, .status' session.json`
 - **Count tasks**: `find .task/ -name "*.json" -type f | wc -l`
 - **Count completed**: `find .summaries/ -name "*.md" -type f 2>/dev/null | wc -l`
@@ -89,11 +89,8 @@ Total: 3 sessions (1 active, 1 paused, 1 completed)
 ### Quick Commands
 ```bash
 # Count all sessions
-ls .workflow/WFS-* | wc -l
-
-# Show only active
-ls .workflow/.active-* | basename | sed 's/^\.active-//'
+ls .workflow/active/WFS-* | wc -l
 
 # Show recent sessions
-ls -t .workflow/WFS-*/workflow-session.json | head -3
+ls -t .workflow/active/WFS-*/workflow-session.json | head -3
 ```

.claude/commands/workflow/session/resume.md

@@ -1,6 +1,6 @@
 ---
 name: resume
-description: Resume the most recently paused workflow session
+description: Resume the most recently paused workflow session with automatic session discovery and status update
 ---
 
 # Resume Workflow Session (/workflow:session:resume)
@@ -17,45 +17,39 @@ Resume the most recently paused workflow session, restoring all context and stat
 
 ### Step 1: Find Paused Sessions
 ```bash
-ls .workflow/WFS-* 2>/dev/null
+ls .workflow/active/WFS-* 2>/dev/null
 ```
 
 ### Step 2: Check Session Status
 ```bash
-jq -r '.status' .workflow/WFS-session/workflow-session.json
+jq -r '.status' .workflow/active/WFS-session/workflow-session.json
 ```
 
 ### Step 3: Find Most Recent Paused
 ```bash
-ls -t .workflow/WFS-*/workflow-session.json | head -1
+ls -t .workflow/active/WFS-*/workflow-session.json | head -1
 ```
 
 ### Step 4: Update Session Status
 ```bash
-jq '.status = "active"' .workflow/WFS-session/workflow-session.json > temp.json
-mv temp.json .workflow/WFS-session/workflow-session.json
+jq '.status = "active"' .workflow/active/WFS-session/workflow-session.json > temp.json
+mv temp.json .workflow/active/WFS-session/workflow-session.json
 ```
 
 ### Step 5: Add Resume Timestamp
 ```bash
-jq '.resumed_at = "'$(date -u +%Y-%m-%dT%H:%M:%SZ)'"' .workflow/WFS-session/workflow-session.json > temp.json
-mv temp.json .workflow/WFS-session/workflow-session.json
-```
-
-### Step 6: Create Active Marker
-```bash
-touch .workflow/.active-WFS-session-name
+jq '.resumed_at = "'$(date -u +%Y-%m-%dT%H:%M:%SZ)'"' .workflow/active/WFS-session/workflow-session.json > temp.json
+mv temp.json .workflow/active/WFS-session/workflow-session.json
 ```
 
 ## Simple Bash Commands
 
 ### Basic Operations
-- **List sessions**: `ls .workflow/WFS-*`
+- **List sessions**: `ls .workflow/active/WFS-*`
 - **Check status**: `jq -r '.status' session.json`
-- **Find recent**: `ls -t .workflow/*/workflow-session.json | head -1`
+- **Find recent**: `ls -t .workflow/active/*/workflow-session.json | head -1`
 - **Update status**: `jq '.status = "active"' session.json > temp.json`
 - **Add timestamp**: `jq '.resumed_at = "'$(date -u +%Y-%m-%dT%H:%M:%SZ)'"'`
-- **Create marker**: `touch .workflow/.active-session`
 
 ### Resume Result
 ```

.claude/commands/workflow/session/solidify.md

@@ -0,0 +1,299 @@
+---
+name: solidify
+description: Crystallize session learnings and user-defined constraints into permanent project guidelines
+argument-hint: "[--type <convention|constraint|learning>] [--category <category>] \"rule or insight\""
+examples:
+  - /workflow:session:solidify "Use functional components for all React code" --type convention
+  - /workflow:session:solidify "No direct DB access from controllers" --type constraint --category architecture
+  - /workflow:session:solidify "Cache invalidation requires event sourcing" --type learning --category architecture
+  - /workflow:session:solidify --interactive
+---
+
+# Session Solidify Command (/workflow:session:solidify)
+
+## Overview
+
+Crystallizes ephemeral session context (insights, decisions, constraints) into permanent project guidelines stored in `.workflow/project-guidelines.json`. This ensures valuable learnings persist across sessions and inform future planning.
+
+## Use Cases
+
+1. **During Session**: Capture important decisions as they're made
+2. **After Session**: Reflect on lessons learned before archiving
+3. **Proactive**: Add team conventions or architectural rules
+
+## Parameters
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| `rule` | string | ✅ (unless --interactive) | The rule, convention, or insight to solidify |
+| `--type` | enum | ❌ | Type: `convention`, `constraint`, `learning` (default: auto-detect) |
+| `--category` | string | ❌ | Category for organization (see categories below) |
+| `--interactive` | flag | ❌ | Launch guided wizard for adding rules |
+
+### Type Categories
+
+**convention** → Coding style preferences (goes to `conventions` section)
+- Subcategories: `coding_style`, `naming_patterns`, `file_structure`, `documentation`
+
+**constraint** → Hard rules that must not be violated (goes to `constraints` section)
+- Subcategories: `architecture`, `tech_stack`, `performance`, `security`
+
+**learning** → Session-specific insights (goes to `learnings` array)
+- Subcategories: `architecture`, `performance`, `security`, `testing`, `process`, `other`
+
+## Execution Process
+
+```
+Input Parsing:
+   ├─ Parse: rule text (required unless --interactive)
+   ├─ Parse: --type (convention|constraint|learning)
+   ├─ Parse: --category (subcategory)
+   └─ Parse: --interactive (flag)
+
+Step 1: Ensure Guidelines File Exists
+   └─ If not exists → Create with empty structure
+
+Step 2: Auto-detect Type (if not specified)
+   └─ Analyze rule text for keywords
+
+Step 3: Validate and Format Entry
+   └─ Build entry object based on type
+
+Step 4: Update Guidelines File
+   └─ Add entry to appropriate section
+
+Step 5: Display Confirmation
+   └─ Show what was added and where
+```
+
+## Implementation
+
+### Step 1: Ensure Guidelines File Exists
+
+```bash
+bash(test -f .workflow/project-guidelines.json && echo "EXISTS" || echo "NOT_FOUND")
+```
+
+**If NOT_FOUND**, create scaffold:
+
+```javascript
+const scaffold = {
+  conventions: {
+    coding_style: [],
+    naming_patterns: [],
+    file_structure: [],
+    documentation: []
+  },
+  constraints: {
+    architecture: [],
+    tech_stack: [],
+    performance: [],
+    security: []
+  },
+  quality_rules: [],
+  learnings: [],
+  _metadata: {
+    created_at: new Date().toISOString(),
+    version: "1.0.0"
+  }
+};
+
+Write('.workflow/project-guidelines.json', JSON.stringify(scaffold, null, 2));
+```
+
+### Step 2: Auto-detect Type (if not specified)
+
+```javascript
+function detectType(ruleText) {
+  const text = ruleText.toLowerCase();
+
+  // Constraint indicators
+  if (/\b(no|never|must not|forbidden|prohibited|always must)\b/.test(text)) {
+    return 'constraint';
+  }
+
+  // Learning indicators
+  if (/\b(learned|discovered|realized|found that|turns out)\b/.test(text)) {
+    return 'learning';
+  }
+
+  // Default to convention
+  return 'convention';
+}
+
+function detectCategory(ruleText, type) {
+  const text = ruleText.toLowerCase();
+
+  if (type === 'constraint' || type === 'learning') {
+    if (/\b(architecture|layer|module|dependency|circular)\b/.test(text)) return 'architecture';
+    if (/\b(security|auth|permission|sanitize|xss|sql)\b/.test(text)) return 'security';
+    if (/\b(performance|cache|lazy|async|sync|slow)\b/.test(text)) return 'performance';
+    if (/\b(test|coverage|mock|stub)\b/.test(text)) return 'testing';
+  }
+
+  if (type === 'convention') {
+    if (/\b(name|naming|prefix|suffix|camel|pascal)\b/.test(text)) return 'naming_patterns';
+    if (/\b(file|folder|directory|structure|organize)\b/.test(text)) return 'file_structure';
+    if (/\b(doc|comment|jsdoc|readme)\b/.test(text)) return 'documentation';
+    return 'coding_style';
+  }
+
+  return type === 'constraint' ? 'tech_stack' : 'other';
+}
+```
+
+### Step 3: Build Entry
+
+```javascript
+function buildEntry(rule, type, category, sessionId) {
+  if (type === 'learning') {
+    return {
+      date: new Date().toISOString().split('T')[0],
+      session_id: sessionId || null,
+      insight: rule,
+      category: category,
+      context: null
+    };
+  }
+
+  // For conventions and constraints, just return the rule string
+  return rule;
+}
+```
+
+### Step 4: Update Guidelines File
+
+```javascript
+const guidelines = JSON.parse(Read('.workflow/project-guidelines.json'));
+
+if (type === 'convention') {
+  if (!guidelines.conventions[category]) {
+    guidelines.conventions[category] = [];
+  }
+  if (!guidelines.conventions[category].includes(rule)) {
+    guidelines.conventions[category].push(rule);
+  }
+} else if (type === 'constraint') {
+  if (!guidelines.constraints[category]) {
+    guidelines.constraints[category] = [];
+  }
+  if (!guidelines.constraints[category].includes(rule)) {
+    guidelines.constraints[category].push(rule);
+  }
+} else if (type === 'learning') {
+  guidelines.learnings.push(buildEntry(rule, type, category, sessionId));
+}
+
+guidelines._metadata.updated_at = new Date().toISOString();
+guidelines._metadata.last_solidified_by = sessionId;
+
+Write('.workflow/project-guidelines.json', JSON.stringify(guidelines, null, 2));
+```
+
+### Step 5: Display Confirmation
+
+```
+✓ Guideline solidified
+
+Type: ${type}
+Category: ${category}
+Rule: "${rule}"
+
+Location: .workflow/project-guidelines.json → ${type}s.${category}
+
+Total ${type}s in ${category}: ${count}
+```
+
+## Interactive Mode
+
+When `--interactive` flag is provided:
+
+```javascript
+AskUserQuestion({
+  questions: [
+    {
+      question: "What type of guideline are you adding?",
+      header: "Type",
+      multiSelect: false,
+      options: [
+        { label: "Convention", description: "Coding style preference (e.g., use functional components)" },
+        { label: "Constraint", description: "Hard rule that must not be violated (e.g., no direct DB access)" },
+        { label: "Learning", description: "Insight from this session (e.g., cache invalidation needs events)" }
+      ]
+    }
+  ]
+});
+
+// Follow-up based on type selection...
+```
+
+## Examples
+
+### Add a Convention
+```bash
+/workflow:session:solidify "Use async/await instead of callbacks" --type convention --category coding_style
+```
+
+Result in `project-guidelines.json`:
+```json
+{
+  "conventions": {
+    "coding_style": ["Use async/await instead of callbacks"]
+  }
+}
+```
+
+### Add an Architectural Constraint
+```bash
+/workflow:session:solidify "No direct DB access from controllers" --type constraint --category architecture
+```
+
+Result:
+```json
+{
+  "constraints": {
+    "architecture": ["No direct DB access from controllers"]
+  }
+}
+```
+
+### Capture a Session Learning
+```bash
+/workflow:session:solidify "Cache invalidation requires event sourcing for consistency" --type learning
+```
+
+Result:
+```json
+{
+  "learnings": [
+    {
+      "date": "2024-12-28",
+      "session_id": "WFS-auth-feature",
+      "insight": "Cache invalidation requires event sourcing for consistency",
+      "category": "architecture"
+    }
+  ]
+}
+```
+
+## Integration with Planning
+
+The `project-guidelines.json` is consumed by:
+
+1. **`/workflow:tools:context-gather`**: Loads guidelines into context-package.json
+2. **`/workflow:plan`**: Passes guidelines to task generation agent
+3. **`task-generate-agent`**: Includes guidelines as "CRITICAL CONSTRAINTS" in system prompt
+
+This ensures all future planning respects solidified rules without users needing to re-state them.
+
+## Error Handling
+
+- **Duplicate Rule**: Warn and skip if exact rule already exists
+- **Invalid Category**: Suggest valid categories for the type
+- **File Corruption**: Backup existing file before modification
+
+## Related Commands
+
+- `/workflow:session:start` - Start a session (may prompt for solidify at end)
+- `/workflow:session:complete` - Complete session (prompts for learnings to solidify)
+- `/workflow:init` - Creates project-guidelines.json scaffold if missing

.claude/commands/workflow/session/start.md

@@ -1,11 +1,13 @@
 ---
 name: start
-description: Discover existing sessions or start a new workflow session with intelligent session management
-argument-hint: [--auto|--new] [optional: task description for new session]
+description: Discover existing sessions or start new workflow session with intelligent session management and conflict detection
+argument-hint: [--type <workflow|review|tdd|test|docs>] [--auto|--new] [optional: task description for new session]
 examples:
   - /workflow:session:start
   - /workflow:session:start --auto "implement OAuth2 authentication"
-  - /workflow:session:start --new "fix login bug"
+  - /workflow:session:start --type review "Code review for auth module"
+  - /workflow:session:start --type tdd --auto "implement user authentication"
+  - /workflow:session:start --type test --new "test payment flow"
 ---
 
 # Start Workflow Session (/workflow:session:start)
@@ -13,6 +15,55 @@ examples:
 ## Overview
 Manages workflow sessions with three operation modes: discovery (manual), auto (intelligent), and force-new.
 
+**Dual Responsibility**:
+1. **Project-level initialization** (first-time only): Creates `.workflow/project.json` for feature registry
+2. **Session-level initialization** (always): Creates session directory structure
+
+## Session Types
+
+The `--type` parameter classifies sessions for CCW dashboard organization:
+
+| Type | Description | Default For |
+|------|-------------|-------------|
+| `workflow` | Standard implementation (default) | `/workflow:plan` |
+| `review` | Code review sessions | `/workflow:review-module-cycle` |
+| `tdd` | TDD-based development | `/workflow:tdd-plan` |
+| `test` | Test generation/fix sessions | `/workflow:test-fix-gen` |
+| `docs` | Documentation sessions | `/memory:docs` |
+
+**Validation**: If `--type` is provided with invalid value, return error:
+```
+ERROR: Invalid session type. Valid types: workflow, review, tdd, test, docs
+```
+
+## Step 0: Initialize Project State (First-time Only)
+
+**Executed before all modes** - Ensures project-level state files exist by calling `/workflow:init`.
+
+### Check and Initialize
+```bash
+# Check if project state exists (both files required)
+bash(test -f .workflow/project-tech.json && echo "TECH_EXISTS" || echo "TECH_NOT_FOUND")
+bash(test -f .workflow/project-guidelines.json && echo "GUIDELINES_EXISTS" || echo "GUIDELINES_NOT_FOUND")
+```
+
+**If either NOT_FOUND**, delegate to `/workflow:init`:
+```javascript
+// Call workflow:init for intelligent project analysis
+SlashCommand({command: "/workflow:init"});
+
+// Wait for init completion
+// project-tech.json and project-guidelines.json will be created
+```
+
+**Output**:
+- If BOTH_EXIST: `PROJECT_STATE: initialized`
+- If NOT_FOUND: Calls `/workflow:init` → creates:
+  - `.workflow/project-tech.json` with full technical analysis
+  - `.workflow/project-guidelines.json` with empty scaffold
+
+**Note**: `/workflow:init` uses cli-explore-agent to build comprehensive project understanding (technology stack, architecture, key components). This step runs once per project. Subsequent executions skip initialization.
+
 ## Mode 1: Discovery Mode (Default)
 
 ### Usage
@@ -20,19 +71,14 @@ Manages workflow sessions with three operation modes: discovery (manual), auto (
 /workflow:session:start
 ```
 
-### Step 1: Check Active Sessions
+### Step 1: List Active Sessions
 ```bash
-bash(ls .workflow/.active-* 2>/dev/null)
+bash(ls -1 .workflow/active/ 2>/dev/null | head -5)
 ```
 
-### Step 2: List All Sessions
+### Step 2: Display Session Metadata
 ```bash
-bash(ls -1 .workflow/WFS-* 2>/dev/null | head -5)
-```
-
-### Step 3: Display Session Metadata
-```bash
-bash(cat .workflow/WFS-promptmaster-platform/workflow-session.json)
+bash(cat .workflow/active/WFS-promptmaster-platform/workflow-session.json)
 ```
 
 ### Step 4: User Decision
@@ -49,7 +95,7 @@ Present session information and wait for user to select or create session.
 
 ### Step 1: Check Active Sessions Count
 ```bash
-bash(ls .workflow/.active-* 2>/dev/null | wc -l)
+bash(find .workflow/active/ -name "WFS-*" -type d 2>/dev/null | wc -l)
 ```
 
 ### Step 2a: No Active Sessions → Create New
@@ -58,15 +104,12 @@ bash(ls .workflow/.active-* 2>/dev/null | wc -l)
 bash(echo "implement OAuth2 auth" | sed 's/[^a-zA-Z0-9]/-/g' | tr '[:upper:]' '[:lower:]' | cut -c1-50)
 
 # Create directory structure
-bash(mkdir -p .workflow/WFS-implement-oauth2-auth/.process)
-bash(mkdir -p .workflow/WFS-implement-oauth2-auth/.task)
-bash(mkdir -p .workflow/WFS-implement-oauth2-auth/.summaries)
+bash(mkdir -p .workflow/active/WFS-implement-oauth2-auth/.process)
+bash(mkdir -p .workflow/active/WFS-implement-oauth2-auth/.task)
+bash(mkdir -p .workflow/active/WFS-implement-oauth2-auth/.summaries)
 
-# Create metadata
-bash(echo '{"session_id":"WFS-implement-oauth2-auth","project":"implement OAuth2 auth","status":"planning"}' > .workflow/WFS-implement-oauth2-auth/workflow-session.json)
-
-# Mark as active
-bash(touch .workflow/.active-WFS-implement-oauth2-auth)
+# Create metadata (include type field, default to "workflow" if not specified)
+bash(echo '{"session_id":"WFS-implement-oauth2-auth","project":"implement OAuth2 auth","status":"planning","type":"workflow","created_at":"2024-12-04T08:00:00Z"}' > .workflow/active/WFS-implement-oauth2-auth/workflow-session.json)
 ```
 
 **Output**: `SESSION_ID: WFS-implement-oauth2-auth`
@@ -74,10 +117,10 @@ bash(touch .workflow/.active-WFS-implement-oauth2-auth)
 ### Step 2b: Single Active Session → Check Relevance
 ```bash
 # Extract session ID
-bash(ls .workflow/.active-* 2>/dev/null | head -1 | xargs basename | sed 's/^\.active-//')
+bash(find .workflow/active/ -name "WFS-*" -type d 2>/dev/null | head -1 | xargs basename)
 
 # Read project name from metadata
-bash(cat .workflow/WFS-promptmaster-platform/workflow-session.json | grep -o '"project":"[^"]*"' | cut -d'"' -f4)
+bash(cat .workflow/active/WFS-promptmaster-platform/workflow-session.json | grep -o '"project":"[^"]*"' | cut -d'"' -f4)
 
 # Check keyword match (manual comparison)
 # If task contains project keywords → Reuse session
@@ -90,7 +133,7 @@ bash(cat .workflow/WFS-promptmaster-platform/workflow-session.json | grep -o '"p
 ### Step 2c: Multiple Active Sessions → Use First
 ```bash
 # Get first active session
-bash(ls .workflow/.active-* 2>/dev/null | head -1 | xargs basename | sed 's/^\.active-//')
+bash(find .workflow/active/ -name "WFS-*" -type d 2>/dev/null | head -1 | xargs basename)
 
 # Output warning and session ID
 # WARNING: Multiple active sessions detected
@@ -110,29 +153,28 @@ bash(ls .workflow/.active-* 2>/dev/null | head -1 | xargs basename | sed 's/^\.a
 bash(echo "fix login bug" | sed 's/[^a-zA-Z0-9]/-/g' | tr '[:upper:]' '[:lower:]' | cut -c1-50)
 
 # Check if exists, add counter if needed
-bash(ls .workflow/WFS-fix-login-bug 2>/dev/null && echo "WFS-fix-login-bug-2" || echo "WFS-fix-login-bug")
+bash(ls .workflow/active/WFS-fix-login-bug 2>/dev/null && echo "WFS-fix-login-bug-2" || echo "WFS-fix-login-bug")
 ```
 
 ### Step 2: Create Session Structure
 ```bash
-bash(mkdir -p .workflow/WFS-fix-login-bug/.process)
-bash(mkdir -p .workflow/WFS-fix-login-bug/.task)
-bash(mkdir -p .workflow/WFS-fix-login-bug/.summaries)
+bash(mkdir -p .workflow/active/WFS-fix-login-bug/.process)
+bash(mkdir -p .workflow/active/WFS-fix-login-bug/.task)
+bash(mkdir -p .workflow/active/WFS-fix-login-bug/.summaries)
 ```
 
 ### Step 3: Create Metadata
 ```bash
-bash(echo '{"session_id":"WFS-fix-login-bug","project":"fix login bug","status":"planning"}' > .workflow/WFS-fix-login-bug/workflow-session.json)
-```
-
-### Step 4: Mark Active and Clean Old Markers
-```bash
-bash(rm .workflow/.active-* 2>/dev/null)
-bash(touch .workflow/.active-WFS-fix-login-bug)
+# Include type field from --type parameter (default: "workflow")
+bash(echo '{"session_id":"WFS-fix-login-bug","project":"fix login bug","status":"planning","type":"workflow","created_at":"2024-12-04T08:00:00Z"}' > .workflow/active/WFS-fix-login-bug/workflow-session.json)
 ```
 
 **Output**: `SESSION_ID: WFS-fix-login-bug`
 
+## Execution Guideline
+
+- **Non-interrupting**: When called from other commands, this command completes and returns control to the caller without interrupting subsequent tasks.
+
 ## Output Format Specification
 
 ### Success
@@ -153,60 +195,6 @@ DECISION: Reusing existing session
 SESSION_ID: WFS-promptmaster-platform
 ```
 
-## Command Integration
-
-### For /workflow:plan (Use Auto Mode)
-```bash
-SlashCommand(command="/workflow:session:start --auto \"implement OAuth2 authentication\"")
-
-# Parse session ID from output
-grep "^SESSION_ID:" | awk '{print $2}'
-```
-
-### For Interactive Workflows (Use Discovery Mode)
-```bash
-SlashCommand(command="/workflow:session:start")
-```
-
-### For New Isolated Work (Use Force New Mode)
-```bash
-SlashCommand(command="/workflow:session:start --new \"experimental feature\"")
-```
-
-## Simple Bash Commands
-
-### Basic Operations
-```bash
-# Check active sessions
-bash(ls .workflow/.active-*)
-
-# List all sessions
-bash(ls .workflow/WFS-*)
-
-# Read session metadata
-bash(cat .workflow/WFS-[session-id]/workflow-session.json)
-
-# Create session directories
-bash(mkdir -p .workflow/WFS-[session-id]/.process)
-bash(mkdir -p .workflow/WFS-[session-id]/.task)
-bash(mkdir -p .workflow/WFS-[session-id]/.summaries)
-
-# Mark session as active
-bash(touch .workflow/.active-WFS-[session-id])
-
-# Clean active markers
-bash(rm .workflow/.active-*)
-```
-
-### Generate Session Slug
-```bash
-bash(echo "Task Description" | sed 's/[^a-zA-Z0-9]/-/g' | tr '[:upper:]' '[:lower:]' | cut -c1-50)
-```
-
-### Create Metadata JSON
-```bash
-bash(echo '{"session_id":"WFS-test","project":"test project","status":"planning"}' > .workflow/WFS-test/workflow-session.json)
-```
 
 ## Session ID Format
 - Pattern: `WFS-[lowercase-slug]`

.claude/commands/workflow/status.md

@@ -1,119 +0,0 @@
----
-name: workflow:status
-description: Generate on-demand views from JSON task data
-argument-hint: "[optional: task-id]"
----
-
-# Workflow Status Command (/workflow:status)
-
-## Overview
-Generates on-demand views from JSON task data. No synchronization needed - all views are calculated from the current state of JSON files.
-
-## Usage
-```bash
-/workflow:status                    # Show current workflow overview
-/workflow:status impl-1             # Show specific task details
-/workflow:status --validate         # Validate workflow integrity
-```
-
-## Implementation Flow
-
-### Step 1: Find Active Session
-```bash
-find .workflow/ -name ".active-*" -type f 2>/dev/null | head -1
-```
-
-### Step 2: Load Session Data
-```bash
-cat .workflow/WFS-session/workflow-session.json
-```
-
-### Step 3: Scan Task Files
-```bash
-find .workflow/WFS-session/.task/ -name "*.json" -type f 2>/dev/null
-```
-
-### Step 4: Generate Task Status
-```bash
-cat .workflow/WFS-session/.task/impl-1.json | jq -r '.status'
-```
-
-### Step 5: Count Task Progress
-```bash
-find .workflow/WFS-session/.task/ -name "*.json" -type f | wc -l
-find .workflow/WFS-session/.summaries/ -name "*.md" -type f 2>/dev/null | wc -l
-```
-
-### Step 6: Display Overview
-```markdown
-# Workflow Overview
-**Session**: WFS-session-name
-**Progress**: 3/8 tasks completed
-
-## Active Tasks
-- [IN PROGRESS] impl-1: Current task in progress
-- [ ] impl-2: Next pending task
-
-## Completed Tasks
-- [COMPLETED] impl-0: Setup completed
-```
-
-## Simple Bash Commands
-
-### Basic Operations
-- **Find active session**: `find .workflow/ -name ".active-*" -type f`
-- **Read session info**: `cat .workflow/session/workflow-session.json`
-- **List tasks**: `find .workflow/session/.task/ -name "*.json" -type f`
-- **Check task status**: `cat task.json | jq -r '.status'`
-- **Count completed**: `find .summaries/ -name "*.md" -type f | wc -l`
-
-### Task Status Check
-- **pending**: Not started yet
-- **active**: Currently in progress
-- **completed**: Finished with summary
-- **blocked**: Waiting for dependencies
-
-### Validation Commands
-```bash
-# Check session exists
-test -f .workflow/.active-* && echo "Session active"
-
-# Validate task files
-for f in .workflow/session/.task/*.json; do jq empty "$f" && echo "Valid: $f"; done
-
-# Check summaries match
-find .task/ -name "*.json" -type f | wc -l
-find .summaries/ -name "*.md" -type f 2>/dev/null | wc -l
-```
-
-## Simple Output Format
-
-### Default Overview
-```
-Session: WFS-user-auth
-Status: ACTIVE
-Progress: 5/12 tasks
-
-Current: impl-3 (Building API endpoints)
-Next: impl-4 (Adding authentication)
-Completed: impl-1, impl-2
-```
-
-### Task Details
-```
-Task: impl-1
-Title: Build authentication module
-Status: completed
-Agent: code-developer
-Created: 2025-09-15
-Completed: 2025-09-15
-Summary: .summaries/impl-1-summary.md
-```
-
-### Validation Results
-```
-Session file valid
-8 task files found
-3 summaries found
-5 tasks pending completion
-```

.claude/commands/workflow/tdd-plan.md

@@ -1,7 +1,7 @@
 ---
 name: tdd-plan
-description: Orchestrate TDD workflow planning with Red-Green-Refactor task chains
-argument-hint: "[--agent] \"feature description\"|file.md"
+description: TDD workflow planning with Red-Green-Refactor task chain generation, test-first development structure, and cycle tracking
+argument-hint: "\"feature description\"|file.md"
 allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 ---
 
@@ -9,26 +9,43 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 
 ## Coordinator Role
 
-**This command is a pure orchestrator**: Execute 5 slash commands in sequence, parse outputs, pass context, and ensure complete TDD workflow creation.
+**This command is a pure orchestrator**: Dispatches 6 slash commands in sequence, parse outputs, pass context, and ensure complete TDD workflow creation with Red-Green-Refactor task generation.
 
-**Execution Modes**:
-- **Manual Mode** (default): Use `/workflow:tools:task-generate-tdd`
-- **Agent Mode** (`--agent`): Use `/workflow:tools:task-generate-tdd --agent`
+**CLI Tool Selection**: CLI tool usage is determined semantically from user's task description. Include "use Codex/Gemini/Qwen" in your request for CLI execution.
+
+**Task Attachment Model**:
+- SlashCommand dispatch **expands workflow** by attaching sub-tasks to current TodoWrite
+- When dispatching a sub-command (e.g., `/workflow:tools:test-context-gather`), its internal tasks are attached to the orchestrator's TodoWrite
+- Orchestrator **executes these attached tasks** sequentially
+- After completion, attached tasks are **collapsed** back to high-level phase summary
+- This is **task expansion**, not external delegation
+
+**Auto-Continue Mechanism**:
+- TodoList tracks current phase status and dynamically manages task attachment/collapse
+- When each phase finishes executing, automatically dispatch next pending phase
+- All phases run autonomously without user interaction
+- **⚠️ CONTINUOUS EXECUTION** - Do not stop until all phases complete
 
 ## Core Rules
 
-1. **Start Immediately**: First action is TodoWrite initialization, second action is Phase 1 execution
+1. **Start Immediately**: First action is TodoWrite initialization, second action is dispatch Phase 1
 2. **No Preliminary Analysis**: Do not read files before Phase 1
 3. **Parse Every Output**: Extract required data for next phase
-4. **Auto-Continue via TodoList**: Check TodoList status to execute next pending phase automatically
-5. **Track Progress**: Update TodoWrite after every phase completion
+4. **Auto-Continue via TodoList**: Check TodoList status to dispatch next pending phase automatically
+5. **Track Progress**: Update TodoWrite dynamically with task attachment/collapse pattern
 6. **TDD Context**: All descriptions include "TDD:" prefix
-7. **Quality Gate**: Phase 4 conflict resolution (optional, auto-triggered) validates compatibility before task generation
+7. **Task Attachment Model**: SlashCommand dispatch **attaches** sub-tasks to current workflow. Orchestrator **executes** these attached tasks itself, then **collapses** them after completion
+8. **⚠️ CRITICAL: DO NOT STOP**: Continuous multi-phase workflow. After executing all attached tasks, immediately collapse them and dispatch next phase
 
 ## 6-Phase Execution (with Conflict Resolution)
 
 ### Phase 1: Session Discovery
-**Command**: `/workflow:session:start --auto "TDD: [structured-description]"`
+
+**Step 1.1: Dispatch** - Session discovery and initialization
+
+```javascript
+SlashCommand(command="/workflow:session:start --type tdd --auto \"TDD: [structured-description]\"")
+```
 
 **TDD Structured Format**:
 ```
@@ -48,7 +65,12 @@ TEST_FOCUS: [Test scenarios]
 ---
 
 ### Phase 2: Context Gathering
-**Command**: `/workflow:tools:context-gather --session [sessionId] "TDD: [structured-description]"`
+
+**Step 2.1: Dispatch** - Context gathering and analysis
+
+```javascript
+SlashCommand(command="/workflow:tools:context-gather --session [sessionId] \"TDD: [structured-description]\"")
+```
 
 **Use Same Structured Description**: Pass the same structured format from Phase 1
 
@@ -56,7 +78,7 @@ TEST_FOCUS: [Test scenarios]
 
 **Parse Output**:
 - Extract: context-package.json path (store as `contextPath`)
-- Typical pattern: `.workflow/[sessionId]/.process/context-package.json`
+- Typical pattern: `.workflow/active/[sessionId]/.process/context-package.json`
 
 **Validation**:
 - Context package path extracted
@@ -69,7 +91,12 @@ TEST_FOCUS: [Test scenarios]
 ---
 
 ### Phase 3: Test Coverage Analysis
-**Command**: `/workflow:tools:test-context-gather --session [sessionId]`
+
+**Step 3.1: Dispatch** - Test coverage analysis and framework detection
+
+```javascript
+SlashCommand(command="/workflow:tools:test-context-gather --session [sessionId]")
+```
 
 **Purpose**: Analyze existing codebase for:
 - Existing test patterns and conventions
@@ -77,17 +104,46 @@ TEST_FOCUS: [Test scenarios]
 - Related components and integration points
 - Test framework detection
 
-**Parse**: Extract testContextPath (`.workflow/[sessionId]/.process/test-context-package.json`)
+**Parse**: Extract testContextPath (`.workflow/active/[sessionId]/.process/test-context-package.json`)
 
-**Benefits**:
-- Makes TDD aware of existing environment
-- Identifies reusable test patterns
-- Prevents duplicate test creation
-- Enables integration with existing tests
 
-**TodoWrite**: Mark phase 3 completed, phase 4 in_progress
 
-**After Phase 3**: Return to user showing test coverage results, then auto-continue to Phase 4
+<!-- TodoWrite: When test-context-gather dispatched, INSERT 3 test-context-gather tasks -->
+
+**TodoWrite Update (Phase 3 SlashCommand dispatched - tasks attached)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 3: Test Coverage Analysis", "status": "in_progress", "activeForm": "Executing test coverage analysis"},
+  {"content": "  → Detect test framework and conventions", "status": "in_progress", "activeForm": "Detecting test framework"},
+  {"content": "  → Analyze existing test coverage", "status": "pending", "activeForm": "Analyzing test coverage"},
+  {"content": "  → Identify coverage gaps", "status": "pending", "activeForm": "Identifying coverage gaps"},
+  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
+  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
+]
+```
+
+**Note**: SlashCommand dispatch **attaches** test-context-gather's 3 tasks. Orchestrator **executes** these tasks.
+
+**Next Action**: Tasks attached → **Execute Phase 3.1-3.3** sequentially
+
+<!-- TodoWrite: After Phase 3 tasks complete, REMOVE Phase 3.1-3.3, restore to orchestrator view -->
+
+**TodoWrite Update (Phase 3 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
+  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
+  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
+]
+```
+
+**Note**: Phase 3 tasks completed and collapsed to summary.
+
+**After Phase 3**: Return to user showing test coverage results, then auto-continue to Phase 4/5 (depending on conflict_risk)
 
 ---
 
@@ -95,7 +151,11 @@ TEST_FOCUS: [Test scenarios]
 
 **Trigger**: Only execute when context-package.json indicates conflict_risk is "medium" or "high"
 
-**Command**: `SlashCommand(command="/workflow:tools:conflict-resolution --session [sessionId] --context [contextPath]")`
+**Step 4.1: Dispatch** - Conflict detection and resolution
+
+```javascript
+SlashCommand(command="/workflow:tools:conflict-resolution --session [sessionId] --context [contextPath]")
+```
 
 **Input**:
 - sessionId from Phase 1
@@ -104,23 +164,64 @@ TEST_FOCUS: [Test scenarios]
 
 **Parse Output**:
 - Extract: Execution status (success/skipped/failed)
-- Verify: CONFLICT_RESOLUTION.md file path (if executed)
+- Verify: conflict-resolution.json file path (if executed)
 
 **Validation**:
-- File `.workflow/[sessionId]/.process/CONFLICT_RESOLUTION.md` exists (if executed)
+- File `.workflow/active/[sessionId]/.process/conflict-resolution.json` exists (if executed)
 
 **Skip Behavior**:
 - If conflict_risk is "none" or "low", skip directly to Phase 5
 - Display: "No significant conflicts detected, proceeding to TDD task generation"
 
-**TodoWrite**: Mark phase 4 completed (if executed) or skipped, phase 5 in_progress
+<!-- TodoWrite: If conflict_risk ≥ medium, INSERT 3 conflict-resolution tasks when dispatched -->
+
+**TodoWrite Update (Phase 4 SlashCommand dispatched - tasks attached, if conflict_risk ≥ medium)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
+  {"content": "Phase 4: Conflict Resolution", "status": "in_progress", "activeForm": "Executing conflict resolution"},
+  {"content": "  → Detect conflicts with CLI analysis", "status": "in_progress", "activeForm": "Detecting conflicts"},
+  {"content": "  → Present conflicts to user", "status": "pending", "activeForm": "Presenting conflicts"},
+  {"content": "  → Apply resolution strategies", "status": "pending", "activeForm": "Applying resolution strategies"},
+  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
+  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
+]
+```
+
+**Note**: SlashCommand dispatch **attaches** conflict-resolution's 3 tasks. Orchestrator **executes** these tasks.
+
+**Next Action**: Tasks attached → **Execute Phase 4.1-4.3** sequentially
+
+<!-- TodoWrite: After Phase 4 tasks complete, REMOVE Phase 4.1-4.3, restore to orchestrator view -->
+
+**TodoWrite Update (Phase 4 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
+  {"content": "Phase 4: Conflict Resolution", "status": "completed", "activeForm": "Executing conflict resolution"},
+  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
+  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
+]
+```
+
+**Note**: Phase 4 tasks completed and collapsed to summary.
 
 **After Phase 4**: Return to user showing conflict resolution results (if executed) and selected strategies, then auto-continue to Phase 5
 
 **Memory State Check**:
 - Evaluate current context window usage and memory state
 - If memory usage is high (>110K tokens or approaching context limits):
-  - **Command**: `SlashCommand(command="/compact")`
+
+  **Step 4.5: Dispatch** - Memory compaction
+
+  ```javascript
+  SlashCommand(command="/compact")
+  ```
+
   - This optimizes memory before proceeding to Phase 5
 - Memory compaction is particularly important after analysis phase which may generate extensive documentation
 - Ensures optimal performance and prevents context overflow
@@ -128,9 +229,14 @@ TEST_FOCUS: [Test scenarios]
 ---
 
 ### Phase 5: TDD Task Generation
-**Command**:
-- Manual: `/workflow:tools:task-generate-tdd --session [sessionId]`
-- Agent: `/workflow:tools:task-generate-tdd --session [sessionId] --agent`
+
+**Step 5.1: Dispatch** - TDD task generation via action-planning-agent
+
+```javascript
+SlashCommand(command="/workflow:tools:task-generate-tdd --session [sessionId]")
+```
+
+**Note**: CLI tool usage is determined semantically from user's task description.
 
 **Parse**: Extract feature count, task count (not chain count - tasks now contain internal TDD cycles)
 
@@ -145,6 +251,41 @@ TEST_FOCUS: [Test scenarios]
 - IMPL_PLAN.md contains workflow_type: "tdd" in frontmatter
 - Task count ≤10 (compliance with task limit)
 
+<!-- TodoWrite: When task-generate-tdd dispatched, INSERT 3 task-generate-tdd tasks -->
+
+**TodoWrite Update (Phase 5 SlashCommand dispatched - tasks attached)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
+  {"content": "Phase 5: TDD Task Generation", "status": "in_progress", "activeForm": "Executing TDD task generation"},
+  {"content": "  → Discovery - analyze TDD requirements", "status": "in_progress", "activeForm": "Analyzing TDD requirements"},
+  {"content": "  → Planning - design Red-Green-Refactor cycles", "status": "pending", "activeForm": "Designing TDD cycles"},
+  {"content": "  → Output - generate IMPL tasks with internal TDD phases", "status": "pending", "activeForm": "Generating TDD tasks"},
+  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
+]
+```
+
+**Note**: SlashCommand dispatch **attaches** task-generate-tdd's 3 tasks. Orchestrator **executes** these tasks. Each generated IMPL task will contain internal Red-Green-Refactor cycle.
+
+**Next Action**: Tasks attached → **Execute Phase 5.1-5.3** sequentially
+
+<!-- TodoWrite: After Phase 5 tasks complete, REMOVE Phase 5.1-5.3, restore to orchestrator view -->
+
+**TodoWrite Update (Phase 5 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
+  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
+  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
+  {"content": "Phase 5: TDD Task Generation", "status": "completed", "activeForm": "Executing TDD task generation"},
+  {"content": "Phase 6: TDD Structure Validation", "status": "in_progress", "activeForm": "Validating TDD structure"}
+]
+```
+
+**Note**: Phase 5 tasks completed and collapsed to summary. Each generated IMPL task contains complete Red-Green-Refactor cycle internally.
+
 ### Phase 6: TDD Structure Validation & Action Plan Verification (RECOMMENDED)
 **Internal validation first, then recommend external verification**
 
@@ -182,9 +323,9 @@ Structure:
 [...]
 
 Plans generated:
-- Unified Implementation Plan: .workflow/[sessionId]/IMPL_PLAN.md
+- Unified Implementation Plan: .workflow/active/[sessionId]/IMPL_PLAN.md
   (includes TDD Implementation Tasks section with workflow_type: "tdd")
-- Task List: .workflow/[sessionId]/TODO_LIST.md
+- Task List: .workflow/active/[sessionId]/TODO_LIST.md
   (with internal TDD phase indicators)
 
 TDD Configuration:
@@ -202,45 +343,85 @@ Quality Gate: Consider running /workflow:action-plan-verify to validate TDD task
 
 ## TodoWrite Pattern
 
-```javascript
-// Initialize (Phase 4 added dynamically after Phase 3 if conflict_risk ≥ medium)
-TodoWrite({todos: [
-  {"content": "Execute session discovery", "status": "in_progress", "activeForm": "Executing session discovery"},
-  {"content": "Execute context gathering", "status": "pending", "activeForm": "Executing context gathering"},
-  {"content": "Execute test coverage analysis", "status": "pending", "activeForm": "Executing test coverage analysis"},
-  // Phase 4 todo added dynamically after Phase 3 if conflict_risk ≥ medium
-  {"content": "Execute TDD task generation", "status": "pending", "activeForm": "Executing TDD task generation"},
-  {"content": "Validate TDD structure", "status": "pending", "activeForm": "Validating TDD structure"}
-]})
+**Core Concept**: Dynamic task attachment and collapse for TDD workflow with test coverage analysis and Red-Green-Refactor cycle generation.
 
-// After Phase 3 (if conflict_risk ≥ medium, insert Phase 4 todo)
-TodoWrite({todos: [
-  {"content": "Execute session discovery", "status": "completed", "activeForm": "Executing session discovery"},
-  {"content": "Execute context gathering", "status": "completed", "activeForm": "Executing context gathering"},
-  {"content": "Execute test coverage analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
-  {"content": "Execute conflict resolution", "status": "in_progress", "activeForm": "Executing conflict resolution"},
-  {"content": "Execute TDD task generation", "status": "pending", "activeForm": "Executing TDD task generation"},
-  {"content": "Validate TDD structure", "status": "pending", "activeForm": "Validating TDD structure"}
-]})
+### Key Principles
 
-// After Phase 3 (if conflict_risk is none/low, skip Phase 4, go directly to Phase 5)
-TodoWrite({todos: [
-  {"content": "Execute session discovery", "status": "completed", "activeForm": "Executing session discovery"},
-  {"content": "Execute context gathering", "status": "completed", "activeForm": "Executing context gathering"},
-  {"content": "Execute test coverage analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
-  {"content": "Execute TDD task generation", "status": "in_progress", "activeForm": "Executing TDD task generation"},
-  {"content": "Validate TDD structure", "status": "pending", "activeForm": "Validating TDD structure"}
-]})
+1. **Task Attachment** (when SlashCommand dispatched):
+   - Sub-command's internal tasks are **attached** to orchestrator's TodoWrite
+   - Example: `/workflow:tools:test-context-gather` attaches 3 sub-tasks (Phase 3.1, 3.2, 3.3)
+   - First attached task marked as `in_progress`, others as `pending`
+   - Orchestrator **executes** these attached tasks sequentially
 
-// After Phase 4 (if executed), continue to Phase 5
-TodoWrite({todos: [
-  {"content": "Execute session discovery", "status": "completed", "activeForm": "Executing session discovery"},
-  {"content": "Execute context gathering", "status": "completed", "activeForm": "Executing context gathering"},
-  {"content": "Execute test coverage analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
-  {"content": "Execute conflict resolution", "status": "completed", "activeForm": "Executing conflict resolution"},
-  {"content": "Execute TDD task generation", "status": "in_progress", "activeForm": "Executing TDD task generation"},
-  {"content": "Validate TDD structure", "status": "pending", "activeForm": "Validating TDD structure"}
-]})
+2. **Task Collapse** (after sub-tasks complete):
+   - Remove detailed sub-tasks from TodoWrite
+   - **Collapse** to high-level phase summary
+   - Example: Phase 3.1-3.3 collapse to "Execute test coverage analysis: completed"
+   - Maintains clean orchestrator-level view
+
+3. **Continuous Execution**:
+   - After collapse, automatically proceed to next pending phase
+   - No user intervention required between phases
+   - TodoWrite dynamically reflects current execution state
+
+**Lifecycle Summary**: Initial pending tasks → Phase dispatched (tasks ATTACHED) → Sub-tasks executed sequentially → Phase completed (tasks COLLAPSED to summary) → Next phase begins (conditional Phase 4 if conflict_risk ≥ medium) → Repeat until all phases complete.
+
+### TDD-Specific Features
+
+- **Phase 3**: Test coverage analysis detects existing patterns and gaps
+- **Phase 5**: Generated IMPL tasks contain internal Red-Green-Refactor cycles
+- **Conditional Phase 4**: Conflict resolution only if conflict_risk ≥ medium
+
+
+
+**Note**: See individual Phase descriptions (Phase 3, 4, 5) for detailed TodoWrite Update examples with full JSON structures.
+
+## Execution Flow Diagram
+
+```
+TDD Workflow Orchestrator
+│
+├─ Phase 1: Session Discovery
+│  └─ /workflow:session:start --auto
+│     └─ Returns: sessionId
+│
+├─ Phase 2: Context Gathering
+│  └─ /workflow:tools:context-gather
+│     └─ Returns: context-package.json path
+│
+├─ Phase 3: Test Coverage Analysis                    ← ATTACHED (3 tasks)
+│  └─ /workflow:tools:test-context-gather
+│     ├─ Phase 3.1: Detect test framework
+│     ├─ Phase 3.2: Analyze existing test coverage
+│     └─ Phase 3.3: Identify coverage gaps
+│     └─ Returns: test-context-package.json           ← COLLAPSED
+│
+├─ Phase 4: Conflict Resolution (conditional)
+│  IF conflict_risk ≥ medium:
+│  └─ /workflow:tools:conflict-resolution             ← ATTACHED (3 tasks)
+│     ├─ Phase 4.1: Detect conflicts with CLI
+│     ├─ Phase 4.2: Present conflicts to user
+│     └─ Phase 4.3: Apply resolution strategies
+│     └─ Returns: conflict-resolution.json            ← COLLAPSED
+│  ELSE:
+│  └─ Skip to Phase 5
+│
+├─ Phase 5: TDD Task Generation                       ← ATTACHED (3 tasks)
+│  └─ /workflow:tools:task-generate-tdd
+│     ├─ Phase 5.1: Discovery - analyze TDD requirements
+│     ├─ Phase 5.2: Planning - design Red-Green-Refactor cycles
+│     └─ Phase 5.3: Output - generate IMPL tasks with internal TDD phases
+│     └─ Returns: IMPL-*.json, IMPL_PLAN.md           ← COLLAPSED
+│        (Each IMPL task contains internal Red-Green-Refactor cycle)
+│
+└─ Phase 6: TDD Structure Validation
+   └─ Internal validation + summary returned
+   └─ Recommend: /workflow:action-plan-verify
+
+Key Points:
+• ← ATTACHED: SlashCommand attaches sub-tasks to orchestrator TodoWrite
+• ← COLLAPSED: Sub-tasks executed and collapsed to phase summary
+• TDD-specific: Each generated IMPL task contains complete Red-Green-Refactor cycle
 ```
 
 ## Input Processing
@@ -258,110 +439,6 @@ Convert user input to TDD-structured format:
 - **Command failure**: Keep phase in_progress, report error
 - **TDD validation failure**: Report incomplete chains or wrong dependencies
 
-## TDD Workflow Enhancements
-
-### Overview
-The TDD workflow has been significantly enhanced by integrating best practices from both traditional `plan --agent` and `test-gen` workflows, creating a hybrid approach that bridges the gap between idealized TDD and real-world development complexity.
-
-### Key Improvements
-
-#### 1. Test Coverage Analysis (Phase 3)
-**Adopted from test-gen workflow**
-
-Before planning TDD tasks, the workflow now analyzes the existing codebase:
-- Detects existing test patterns and conventions
-- Identifies current test coverage
-- Discovers related components and integration points
-- Detects test framework automatically
-
-**Benefits**:
-- Context-aware TDD planning
-- Avoids duplicate test creation
-- Enables integration with existing tests
-- No longer assumes greenfield scenarios
-
-#### 2. Iterative Green Phase with Test-Fix Cycle
-**Adopted from test-gen workflow**
-
-IMPL (Green phase) tasks now include automatic test-fix cycle for resilient implementation:
-
-**Enhanced IMPL Task Flow**:
-```
-1. Write minimal implementation code
-2. Execute test suite
-3. IF tests pass → Complete task
-4. IF tests fail → Enter fix cycle:
-   a. Gemini diagnoses with bug-fix template
-   b. Apply fix (manual or Codex)
-   c. Retest
-   d. Repeat (max 3 iterations)
-5. IF max iterations → Auto-revert changes 🔄
-```
-
-**Benefits**:
-- Faster feedback within Green phase
-- Autonomous recovery from implementation errors
-- Systematic debugging with Gemini
-- Safe rollback prevents broken state
-
-#### 3. Agent-Driven Planning
-**From plan --agent workflow**
-
-Supports action-planning-agent for more autonomous TDD planning with:
-- MCP tool integration (code-index, exa)
-- Memory-first principles
-- Brainstorming artifact integration
-- Task merging over decomposition
-
-### Workflow Comparison
-
-| Aspect | Previous | Current (Optimized) |
-|--------|----------|---------------------|
-| **Phases** | 6 (with test coverage) | 7 (added concept verification) |
-| **Context** | Greenfield assumption | Existing codebase aware |
-| **Task Structure** | 1 feature = 3 tasks (TEST/IMPL/REFACTOR) | 1 feature = 1 task (internal TDD cycle) |
-| **Task Count** | 5 features = 15 tasks | 5 features = 5 tasks (70% reduction) |
-| **Green Phase** | Single implementation | Iterative with fix cycle |
-| **Failure Handling** | Manual intervention | Auto-diagnose + fix + revert |
-| **Test Analysis** | None | Deep coverage analysis |
-| **Feedback Loop** | Post-execution | During Green phase |
-| **Task Management** | High overhead (15 tasks) | Low overhead (5 tasks) |
-| **Execution Efficiency** | Frequent context switching | Continuous context per feature |
-
-### Migration Notes
-
-**Backward Compatibility**: Fully compatible
-- Existing TDD workflows continue to work
-- New features are additive, not breaking
-- Phase 3 can be skipped if test-context-gather not available
-
-**Session Structure**:
-```
-.workflow/WFS-xxx/
-├── IMPL_PLAN.md (unified plan with TDD Implementation Tasks section)
-├── TODO_LIST.md (with internal TDD phase indicators)
-├── .process/
-│   ├── context-package.json
-│   ├── test-context-package.json
-│   ├── ANALYSIS_RESULTS.md (enhanced with TDD breakdown)
-│   └── green-fix-iteration-*.md (fix logs from Green phase cycles)
-└── .task/
-    ├── IMPL-1.json (Complete TDD task: Red-Green-Refactor internally)
-    ├── IMPL-2.json (Complete TDD task)
-    ├── IMPL-3.json (Complex feature container, if needed)
-    ├── IMPL-3.1.json (Complex feature subtask, if needed)
-    └── IMPL-3.2.json (Complex feature subtask, if needed)
-```
-
-**File Count Comparison**:
-- **Old structure**: 5 features = 15 task files (TEST/IMPL/REFACTOR × 5)
-- **New structure**: 5 features = 5 task files (IMPL-N × 5)
-- **Complex features**: Add container + subtasks only when necessary
-
-**Configuration Options** (in IMPL tasks):
-- `meta.max_iterations`: Fix attempts (default: 3)
-- `meta.use_codex`: Auto-fix mode (default: false)
-
 ## Related Commands
 
 **Prerequisite Commands**:
@@ -373,8 +450,7 @@ Supports action-planning-agent for more autonomous TDD planning with:
 - `/workflow:tools:test-context-gather` - Phase 3: Analyze existing test patterns and coverage
 - `/workflow:tools:conflict-resolution` - Phase 4: Detect and resolve conflicts (auto-triggered if conflict_risk ≥ medium)
 - `/compact` - Phase 4: Memory optimization (if context approaching limits)
-- `/workflow:tools:task-generate-tdd` - Phase 5: Generate TDD task chains with Red-Green-Refactor cycles
-- `/workflow:tools:task-generate-tdd --agent` - Phase 5: Generate TDD tasks with agent-driven approach (when `--agent` flag used)
+- `/workflow:tools:task-generate-tdd` - Phase 5: Generate TDD tasks (CLI tool usage determined semantically)
 
 **Follow-up Commands**:
 - `/workflow:action-plan-verify` - Recommended: Verify TDD plan quality and structure before execution

.claude/commands/workflow/tdd-verify.md

@@ -1,6 +1,6 @@
 ---
 name: tdd-verify
-description: Verify TDD workflow compliance and generate quality report
+description: Verify TDD workflow compliance against Red-Green-Refactor cycles, generate quality report with coverage analysis
 
 argument-hint: "[optional: WFS-session-id]"
 allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(gemini:*)
@@ -18,6 +18,39 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(gemini:*)
 - Validate TDD cycle execution
 - Generate compliance report
 
+## Execution Process
+
+```
+Input Parsing:
+   └─ Decision (session argument):
+      ├─ session-id provided → Use provided session
+      └─ No session-id → Auto-detect active session
+
+Phase 1: Session Discovery
+   ├─ Validate session directory exists
+   └─ TodoWrite: Mark phase 1 completed
+
+Phase 2: Task Chain Validation
+   ├─ Load all task JSONs from .task/
+   ├─ Extract task IDs and group by feature
+   ├─ Validate TDD structure:
+   │  ├─ TEST-N.M → IMPL-N.M → REFACTOR-N.M chain
+   │  ├─ Dependency verification
+   │  └─ Meta field validation (tdd_phase, agent)
+   └─ TodoWrite: Mark phase 2 completed
+
+Phase 3: Test Execution Analysis
+   └─ /workflow:tools:tdd-coverage-analysis
+      ├─ Coverage metrics extraction
+      ├─ TDD cycle verification
+      └─ Compliance score calculation
+
+Phase 4: Compliance Report Generation
+   ├─ Gemini analysis for comprehensive report
+   ├─ Generate TDD_COMPLIANCE_REPORT.md
+   └─ Return summary to user
+```
+
 ## 4-Phase Execution
 
 ### Phase 1: Session Discovery
@@ -28,7 +61,7 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(gemini:*)
 sessionId = argument
 
 # Else auto-detect active session
-find .workflow/ -name '.active-*' | head -1 | sed 's/.*active-//'
+find .workflow/active/ -name "WFS-*" -type d | head -1 | sed 's/.*\///'
 ```
 
 **Extract**: sessionId
@@ -44,18 +77,32 @@ find .workflow/ -name '.active-*' | head -1 | sed 's/.*active-//'
 
 ```bash
 # Load all task JSONs
-find .workflow/{sessionId}/.task/ -name '*.json'
+for task_file in .workflow/active/{sessionId}/.task/*.json; do
+  cat "$task_file"
+done
 
 # Extract task IDs
-find .workflow/{sessionId}/.task/ -name '*.json' -exec jq -r '.id' {} \;
+for task_file in .workflow/active/{sessionId}/.task/*.json; do
+  cat "$task_file" | jq -r '.id'
+done
 
-# Check dependencies
-find .workflow/{sessionId}/.task/ -name 'IMPL-*.json' -exec jq -r '.context.depends_on[]?' {} \;
-find .workflow/{sessionId}/.task/ -name 'REFACTOR-*.json' -exec jq -r '.context.depends_on[]?' {} \;
+# Check dependencies - read tasks and filter for IMPL/REFACTOR
+for task_file in .workflow/active/{sessionId}/.task/IMPL-*.json; do
+  cat "$task_file" | jq -r '.context.depends_on[]?'
+done
+
+for task_file in .workflow/active/{sessionId}/.task/REFACTOR-*.json; do
+  cat "$task_file" | jq -r '.context.depends_on[]?'
+done
 
 # Check meta fields
-find .workflow/{sessionId}/.task/ -name '*.json' -exec jq -r '.meta.tdd_phase' {} \;
-find .workflow/{sessionId}/.task/ -name '*.json' -exec jq -r '.meta.agent' {} \;
+for task_file in .workflow/active/{sessionId}/.task/*.json; do
+  cat "$task_file" | jq -r '.meta.tdd_phase'
+done
+
+for task_file in .workflow/active/{sessionId}/.task/*.json; do
+  cat "$task_file" | jq -r '.meta.agent'
+done
 ```
 
 **Validation**:
@@ -82,9 +129,9 @@ find .workflow/{sessionId}/.task/ -name '*.json' -exec jq -r '.meta.agent' {} \;
 - Compliance score
 
 **Validation**:
-- `.workflow/{sessionId}/.process/test-results.json` exists
-- `.workflow/{sessionId}/.process/coverage-report.json` exists
-- `.workflow/{sessionId}/.process/tdd-cycle-report.md` exists
+- `.workflow/active/{sessionId}/.process/test-results.json` exists
+- `.workflow/active/{sessionId}/.process/coverage-report.json` exists
+- `.workflow/active/{sessionId}/.process/tdd-cycle-report.md` exists
 
 **TodoWrite**: Mark phase 3 completed, phase 4 in_progress
 
@@ -94,10 +141,10 @@ find .workflow/{sessionId}/.task/ -name '*.json' -exec jq -r '.meta.agent' {} \;
 **Gemini analysis for comprehensive TDD compliance report**
 
 ```bash
-cd project-root && gemini -p "
+ccw cli -p "
 PURPOSE: Generate TDD compliance report
 TASK: Analyze TDD workflow execution and generate quality report
-CONTEXT: @{.workflow/{sessionId}/.task/*.json,.workflow/{sessionId}/.summaries/*,.workflow/{sessionId}/.process/tdd-cycle-report.md}
+CONTEXT: @{.workflow/active/{sessionId}/.task/*.json,.workflow/active/{sessionId}/.summaries/*,.workflow/active/{sessionId}/.process/tdd-cycle-report.md}
 EXPECTED:
 - TDD compliance score (0-100)
 - Chain completeness verification
@@ -106,7 +153,7 @@ EXPECTED:
 - Red-Green-Refactor cycle validation
 - Best practices adherence assessment
 RULES: Focus on TDD best practices and workflow adherence. Be specific about violations and improvements.
-" > .workflow/{sessionId}/TDD_COMPLIANCE_REPORT.md
+" --tool gemini --mode analysis --cd project-root > .workflow/active/{sessionId}/TDD_COMPLIANCE_REPORT.md
 ```
 
 **Output**: TDD_COMPLIANCE_REPORT.md
@@ -134,7 +181,7 @@ Function Coverage: {percentage}%
 
 ## Compliance Score: {score}/100
 
-Detailed report: .workflow/{sessionId}/TDD_COMPLIANCE_REPORT.md
+Detailed report: .workflow/active/{sessionId}/TDD_COMPLIANCE_REPORT.md
 
 Recommendations:
 - Complete missing REFACTOR-3.1 task
@@ -168,7 +215,7 @@ TodoWrite({todos: [
 
 ### Chain Validation Algorithm
 ```
-1. Load all task JSONs from .workflow/{sessionId}/.task/
+1. Load all task JSONs from .workflow/active/{sessionId}/.task/
 2. Extract task IDs and group by feature number
 3. For each feature:
    - Check TEST-N.M exists
@@ -202,7 +249,7 @@ Final Score: Max(0, Base Score - Deductions)
 
 ## Output Files
 ```
-.workflow/{session-id}/
+.workflow/active/{session-id}/
 ├── TDD_COMPLIANCE_REPORT.md     # Comprehensive compliance report ⭐
 └── .process/
     ├── test-results.json         # From tdd-coverage-analysis
@@ -215,8 +262,8 @@ Final Score: Max(0, Base Score - Deductions)
 ### Session Discovery Errors
 | Error | Cause | Resolution |
 |-------|-------|------------|
-| No active session | No .active-* file | Provide session-id explicitly |
-| Multiple active sessions | Multiple .active-* files | Provide session-id explicitly |
+| No active session | No WFS-* directories | Provide session-id explicitly |
+| Multiple active sessions | Multiple WFS-* directories | Provide session-id explicitly |
 | Session not found | Invalid session-id | Check available sessions |
 
 ### Validation Errors

.claude/commands/workflow/test-fix-gen.md

@@ -1,7 +1,7 @@
 ---
 name: test-fix-gen
-description: Create independent test-fix workflow session from existing implementation (session or prompt-based)
-argument-hint: "[--use-codex] [--cli-execute] (source-session-id | \"feature description\" | /path/to/file.md)"
+description: Create test-fix workflow session from session ID, description, or file path with test strategy generation and task planning
+argument-hint: "(source-session-id | \"feature description\" | /path/to/file.md)"
 allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 ---
 
@@ -43,7 +43,7 @@ fi
 - **Session Isolation**: Creates independent `WFS-test-[slug]` session
 - **Context-First**: Gathers implementation context via appropriate method
 - **Format Reuse**: Creates standard `IMPL-*.json` tasks with `meta.type: "test-fix"`
-- **Manual First**: Default to manual fixes, use `--use-codex` for automation
+- **Semantic CLI Selection**: CLI tool usage determined from user's task description
 - **Automatic Detection**: Input pattern determines execution mode
 
 ### Coordinator Role
@@ -58,6 +58,19 @@ This command is a **pure planning coordinator**:
 - Creates independent test workflow session
 - **All execution delegated to `/workflow:test-cycle-execute`**
 
+**Task Attachment Model**:
+- SlashCommand dispatch **expands workflow** by attaching sub-tasks to current TodoWrite
+- When dispatching a sub-command (e.g., `/workflow:tools:test-context-gather`), its internal tasks are attached to the orchestrator's TodoWrite
+- Orchestrator **executes these attached tasks** sequentially
+- After completion, attached tasks are **collapsed** back to high-level phase summary
+- This is **task expansion**, not external delegation
+
+**Auto-Continue Mechanism**:
+- TodoList tracks current phase status and dynamically manages task attachment/collapse
+- When each phase finishes executing, automatically execute next pending phase
+- All phases run autonomously without user interaction
+- **⚠️ CONTINUOUS EXECUTION** - Do not stop until all phases complete
+
 ---
 
 ## Usage
@@ -66,16 +79,14 @@ This command is a **pure planning coordinator**:
 
 ```bash
 # Basic syntax
-/workflow:test-fix-gen [FLAGS] <INPUT>
-
-# Flags (optional)
---use-codex        # Enable Codex automated fixes in IMPL-002
---cli-execute      # Enable CLI execution in IMPL-001
+/workflow:test-fix-gen <INPUT>
 
 # Input
 <INPUT>            # Session ID, description, or file path
 ```
 
+**Note**: CLI tool usage is determined semantically from the task description. To request CLI execution, include it in your description (e.g., "use Codex for automated fixes").
+
 ### Usage Examples
 
 #### Session Mode
@@ -83,11 +94,8 @@ This command is a **pure planning coordinator**:
 # Test validation for completed implementation
 /workflow:test-fix-gen WFS-user-auth-v2
 
-# With automated fixes
-/workflow:test-fix-gen --use-codex WFS-api-endpoints
-
-# With CLI execution
-/workflow:test-fix-gen --cli-execute --use-codex WFS-payment-flow
+# With semantic CLI request
+/workflow:test-fix-gen WFS-api-endpoints  # Add "use Codex" in description for automated fixes
 ```
 
 #### Prompt Mode - Text Description
@@ -95,17 +103,14 @@ This command is a **pure planning coordinator**:
 # Generate tests from feature description
 /workflow:test-fix-gen "Test the user authentication API endpoints in src/auth/api.ts"
 
-# With automated fixes
-/workflow:test-fix-gen --use-codex "Test user registration and login flows"
+# With CLI execution (semantic)
+/workflow:test-fix-gen "Test user registration and login flows, use Codex for automated fixes"
 ```
 
 #### Prompt Mode - File Reference
 ```bash
 # Generate tests from requirements file
 /workflow:test-fix-gen ./docs/api-requirements.md
-
-# With flags
-/workflow:test-fix-gen --use-codex --cli-execute ./specs/feature.md
 ```
 
 ### Mode Comparison
@@ -123,30 +128,50 @@ This command is a **pure planning coordinator**:
 
 ### Core Execution Rules
 
-1. **Start Immediately**: First action is TodoWrite, second is Phase 1 session creation
+1. **Start Immediately**: First action is TodoWrite, second is dispatch Phase 1 session creation
 2. **No Preliminary Analysis**: Do not read files before Phase 1
 3. **Parse Every Output**: Extract required data from each phase for next phase
 4. **Sequential Execution**: Each phase depends on previous phase's output
 5. **Complete All Phases**: Do not return until Phase 5 completes
-6. **Track Progress**: Update TodoWrite after every phase
+6. **Track Progress**: Update TodoWrite dynamically with task attachment/collapse pattern
 7. **Automatic Detection**: Mode auto-detected from input pattern
-8. **Parse Flags**: Extract `--use-codex` and `--cli-execute` flags for Phase 4
+8. **Semantic CLI Detection**: CLI tool usage determined from user's task description for Phase 4
+9. **Task Attachment Model**: SlashCommand dispatch **attaches** sub-tasks to current workflow. Orchestrator **executes** these attached tasks itself, then **collapses** them after completion
+10. **⚠️ CRITICAL: DO NOT STOP**: Continuous multi-phase workflow. After executing all attached tasks, immediately collapse them and execute next phase
 
 ### 5-Phase Execution
 
 #### Phase 1: Create Test Session
 
-**Command**:
-- **Session Mode**: `SlashCommand("/workflow:session:start --new \"Test validation for [sourceSessionId]\"")`
-- **Prompt Mode**: `SlashCommand("/workflow:session:start --new \"Test generation for: [description]\"")`
+**Step 1.0: Load Source Session Intent (Session Mode Only)** - Preserve user's original task description for semantic CLI selection
+
+```javascript
+// Session Mode: Read source session metadata to get original task description
+Read(".workflow/active/[sourceSessionId]/workflow-session.json")
+// OR if context-package exists:
+Read(".workflow/active/[sourceSessionId]/.process/context-package.json")
+
+// Extract: metadata.task_description or project/description field
+// This preserves user's CLI tool preferences (e.g., "use Codex for fixes")
+```
+
+**Step 1.1: Dispatch** - Create test workflow session with preserved intent
+
+```javascript
+// Session Mode - Include original task description to enable semantic CLI selection
+SlashCommand(command="/workflow:session:start --type test --new \"Test validation for [sourceSessionId]: [originalTaskDescription]\"")
+
+// Prompt Mode - User's description already contains their intent
+SlashCommand(command="/workflow:session:start --type test --new \"Test generation for: [description]\"")
+```
 
 **Input**: User argument (session ID, description, or file path)
 
 **Expected Behavior**:
 - Creates new session: `WFS-test-[slug]`
-- Writes `workflow-session.json` metadata:
-  - **Session Mode**: Includes `workflow_type: "test_session"`, `source_session_id: "[sourceId]"`
-  - **Prompt Mode**: Includes `workflow_type: "test_session"` only
+- Writes `workflow-session.json` metadata with `type: "test"`
+  - **Session Mode**: Additionally includes `source_session_id: "[sourceId]"`, description with original user intent
+  - **Prompt Mode**: Uses user's description (already contains intent)
 - Returns new session ID
 
 **Parse Output**:
@@ -162,9 +187,15 @@ This command is a **pure planning coordinator**:
 
 #### Phase 2: Gather Test Context
 
-**Command**:
-- **Session Mode**: `SlashCommand("/workflow:tools:test-context-gather --session [testSessionId]")`
-- **Prompt Mode**: `SlashCommand("/workflow:tools:context-gather --session [testSessionId] \"[task_description]\"")`
+**Step 2.1: Dispatch** - Gather test context via appropriate method
+
+```javascript
+// Session Mode
+SlashCommand(command="/workflow:tools:test-context-gather --session [testSessionId]")
+
+// Prompt Mode
+SlashCommand(command="/workflow:tools:context-gather --session [testSessionId] \"[task_description]\"")
+```
 
 **Input**: `testSessionId` from Phase 1
 
@@ -193,7 +224,11 @@ This command is a **pure planning coordinator**:
 
 #### Phase 3: Test Generation Analysis
 
-**Command**: `SlashCommand("/workflow:tools:test-concept-enhanced --session [testSessionId] --context [contextPath]")`
+**Step 3.1: Dispatch** - Generate test requirements using Gemini
+
+```javascript
+SlashCommand(command="/workflow:tools:test-concept-enhanced --session [testSessionId] --context [contextPath]")
+```
 
 **Input**:
 - `testSessionId` from Phase 1
@@ -202,9 +237,25 @@ This command is a **pure planning coordinator**:
 **Expected Behavior**:
 - Use Gemini to analyze coverage gaps and implementation
 - Study existing test patterns and conventions
-- Generate test requirements for missing test files
-- Design test generation strategy
-- Generate `TEST_ANALYSIS_RESULTS.md`
+- Generate **multi-layered test requirements** (L0: Static Analysis, L1: Unit, L2: Integration, L3: E2E)
+- Design test generation strategy with quality assurance criteria
+- Generate `TEST_ANALYSIS_RESULTS.md` with structured test layers
+
+**Enhanced Test Requirements**:
+For each targeted file/function, Gemini MUST generate:
+1. **L0: Static Analysis Requirements**:
+   - Linting rules to enforce (ESLint, Prettier)
+   - Type checking requirements (TypeScript)
+   - Anti-pattern detection rules
+2. **L1: Unit Test Requirements**:
+   - Happy path scenarios (valid inputs → expected outputs)
+   - Negative path scenarios (invalid inputs → error handling)
+   - Edge cases (null, undefined, 0, empty strings/arrays)
+3. **L2: Integration Test Requirements**:
+   - Successful component interactions
+   - Failure handling scenarios (service unavailable, timeout)
+4. **L3: E2E Test Requirements** (if applicable):
+   - Key user journeys from start to finish
 
 **Parse Output**:
 - Verify `.workflow/[testSessionId]/.process/TEST_ANALYSIS_RESULTS.md` created
@@ -213,9 +264,18 @@ This command is a **pure planning coordinator**:
 - TEST_ANALYSIS_RESULTS.md exists with complete sections:
   - Coverage Assessment
   - Test Framework & Conventions
-  - Test Requirements by File
+  - **Multi-Layered Test Plan** (NEW):
+    - L0: Static Analysis Plan
+    - L1: Unit Test Plan
+    - L2: Integration Test Plan
+    - L3: E2E Test Plan (if applicable)
+  - Test Requirements by File (with layer annotations)
   - Test Generation Strategy
   - Implementation Targets
+  - Quality Assurance Criteria (NEW):
+    - Minimum coverage thresholds
+    - Required test types per function
+    - Acceptance criteria for test quality
   - Success Criteria
 
 **TodoWrite**: Mark phase 3 completed, phase 4 in_progress
@@ -224,24 +284,30 @@ This command is a **pure planning coordinator**:
 
 #### Phase 4: Generate Test Tasks
 
-**Command**: `SlashCommand("/workflow:tools:test-task-generate [--use-codex] [--cli-execute] --session [testSessionId]")`
+**Step 4.1: Dispatch** - Generate test task JSONs
+
+```javascript
+SlashCommand(command="/workflow:tools:test-task-generate --session [testSessionId]")
+```
 
 **Input**:
 - `testSessionId` from Phase 1
-- `--use-codex` flag (if present) - Controls IMPL-002 fix mode
-- `--cli-execute` flag (if present) - Controls IMPL-001 generation mode
+
+**Note**: CLI tool usage is determined semantically from user's task description.
 
 **Expected Behavior**:
-- Parse TEST_ANALYSIS_RESULTS.md from Phase 3
-- Generate **minimum 2 task JSON files** (expandable based on complexity):
+- Parse TEST_ANALYSIS_RESULTS.md from Phase 3 (multi-layered test plan)
+- Generate **minimum 3 task JSON files** (expandable based on complexity):
   - **IMPL-001.json**: Test Understanding & Generation (`@code-developer`)
+  - **IMPL-001.5-review.json**: Test Quality Gate (`@test-fix-agent`) ← **NEW**
   - **IMPL-002.json**: Test Execution & Fix Cycle (`@test-fix-agent`)
   - **IMPL-003+**: Additional tasks if needed for complex projects
-- Generate `IMPL_PLAN.md` with test strategy
+- Generate `IMPL_PLAN.md` with multi-layered test strategy
 - Generate `TODO_LIST.md` with task checklist
 
 **Parse Output**:
 - Verify `.workflow/[testSessionId]/.task/IMPL-001.json` exists
+- Verify `.workflow/[testSessionId]/.task/IMPL-001.5-review.json` exists ← **NEW**
 - Verify `.workflow/[testSessionId]/.task/IMPL-002.json` exists
 - Verify additional `.task/IMPL-*.json` if applicable
 - Verify `IMPL_PLAN.md` and `TODO_LIST.md` created
@@ -262,11 +328,16 @@ Test Session: [testSessionId]
 
 Tasks Created:
 - IMPL-001: Test Understanding & Generation (@code-developer)
+- IMPL-001.5: Test Quality Gate - Static Analysis & Coverage (@test-fix-agent) ← NEW
 - IMPL-002: Test Execution & Fix Cycle (@test-fix-agent)
 [- IMPL-003+: Additional tasks if applicable]
 
+Test Strategy: Multi-Layered (L0: Static, L1: Unit, L2: Integration, L3: E2E)
 Test Framework: [detected framework]
 Test Files to Generate: [count]
+Quality Thresholds:
+- Minimum Coverage: 80%
+- Static Analysis: Zero critical issues
 Max Fix Iterations: 5
 Fix Mode: [Manual|Codex Automated]
 
@@ -275,11 +346,12 @@ Review artifacts:
 - Task list: .workflow/[testSessionId]/TODO_LIST.md
 
 CRITICAL - Next Steps:
-1. Review IMPL_PLAN.md
+1. Review IMPL_PLAN.md (now includes multi-layered test strategy)
 2. **MUST execute: /workflow:test-cycle-execute**
    - This command only generated task JSON files
    - Test execution and fix iterations happen in test-cycle-execute
    - Do NOT attempt to run tests or fixes in main workflow
+3. IMPL-001.5 will validate test quality before fix cycle begins
 ```
 
 **TodoWrite**: Mark phase 5 completed
@@ -291,52 +363,155 @@ CRITICAL - Next Steps:
 
 ---
 
-### TodoWrite Progress Tracking
+### TodoWrite Pattern
 
-Track all 5 phases:
+**Core Concept**: Dynamic task attachment and collapse for test-fix-gen workflow with dual-mode support (Session Mode and Prompt Mode).
 
-```javascript
-TodoWrite({todos: [
-  {"content": "Create independent test session", "status": "in_progress|completed", "activeForm": "Creating test session"},
-  {"content": "Gather test coverage context", "status": "pending|in_progress|completed", "activeForm": "Gathering test coverage context"},
-  {"content": "Analyze test requirements with Gemini", "status": "pending|in_progress|completed", "activeForm": "Analyzing test requirements"},
-  {"content": "Generate test generation and execution tasks", "status": "pending|in_progress|completed", "activeForm": "Generating test tasks"},
-  {"content": "Return workflow summary", "status": "pending|in_progress|completed", "activeForm": "Returning workflow summary"}
-]})
+#### Initial TodoWrite Structure
+
+```json
+[
+  {"content": "Phase 1: Create Test Session", "status": "in_progress", "activeForm": "Creating test session"},
+  {"content": "Phase 2: Gather Test Context", "status": "pending", "activeForm": "Gathering test context"},
+  {"content": "Phase 3: Test Generation Analysis", "status": "pending", "activeForm": "Analyzing test generation"},
+  {"content": "Phase 4: Generate Test Tasks", "status": "pending", "activeForm": "Generating test tasks"},
+  {"content": "Phase 5: Return Summary", "status": "pending", "activeForm": "Completing"}
+]
 ```
 
-Update status to `in_progress` when starting each phase, `completed` when done.
+#### Key Principles
+
+1. **Task Attachment** (when SlashCommand dispatched):
+   - Sub-command's internal tasks are **attached** to orchestrator's TodoWrite
+   - Example - Phase 2 with sub-tasks:
+   ```json
+   [
+     {"content": "Phase 1: Create Test Session", "status": "completed", "activeForm": "Creating test session"},
+     {"content": "Phase 2: Gather Test Context", "status": "in_progress", "activeForm": "Gathering test context"},
+     {"content": "  → Load context and analyze coverage", "status": "in_progress", "activeForm": "Loading context"},
+     {"content": "  → Detect test framework and conventions", "status": "pending", "activeForm": "Detecting framework"},
+     {"content": "  → Generate context package", "status": "pending", "activeForm": "Generating context"},
+     {"content": "Phase 3: Test Generation Analysis", "status": "pending", "activeForm": "Analyzing test generation"},
+     {"content": "Phase 4: Generate Test Tasks", "status": "pending", "activeForm": "Generating test tasks"},
+     {"content": "Phase 5: Return Summary", "status": "pending", "activeForm": "Completing"}
+   ]
+   ```
+
+2. **Task Collapse** (after sub-tasks complete):
+   - Remove detailed sub-tasks from TodoWrite
+   - **Collapse** to high-level phase summary
+   - Example - Phase 2 completed:
+   ```json
+   [
+     {"content": "Phase 1: Create Test Session", "status": "completed", "activeForm": "Creating test session"},
+     {"content": "Phase 2: Gather Test Context", "status": "completed", "activeForm": "Gathering test context"},
+     {"content": "Phase 3: Test Generation Analysis", "status": "in_progress", "activeForm": "Analyzing test generation"},
+     {"content": "Phase 4: Generate Test Tasks", "status": "pending", "activeForm": "Generating test tasks"},
+     {"content": "Phase 5: Return Summary", "status": "pending", "activeForm": "Completing"}
+   ]
+   ```
+
+3. **Continuous Execution**:
+   - After collapse, automatically proceed to next pending phase
+   - No user intervention required between phases
+   - TodoWrite dynamically reflects current execution state
+
+**Lifecycle Summary**: Initial pending tasks → Phase dispatched (tasks ATTACHED with mode-specific context gathering) → Sub-tasks executed sequentially → Phase completed (tasks COLLAPSED to summary) → Next phase begins → Repeat until all phases complete.
+
+#### Test-Fix-Gen Specific Features
+
+- **Dual-Mode Support**: Automatic mode detection based on input pattern
+  - **Session Mode**: Input pattern `WFS-*` → uses `test-context-gather` for cross-session context
+  - **Prompt Mode**: Text or file path → uses `context-gather` for direct codebase analysis
+- **Phase 2**: Mode-specific context gathering (session summaries vs codebase analysis)
+- **Phase 3**: Multi-layered test requirements analysis (L0: Static, L1: Unit, L2: Integration, L3: E2E)
+- **Phase 4**: Multi-task generation with quality gate (IMPL-001, IMPL-001.5-review, IMPL-002)
+- **Fix Mode Configuration**: CLI tool usage determined semantically from user's task description
+
 
 ---
 
 ## Task Specifications
 
-Generates minimum 2 tasks (expandable for complex projects):
+Generates minimum 3 tasks (expandable for complex projects):
 
 ### IMPL-001: Test Understanding & Generation
 
 **Agent**: `@code-developer`
 
-**Purpose**: Understand source implementation and generate test files
+**Purpose**: Understand source implementation and generate test files following multi-layered test strategy
 
 **Task Configuration**:
 - Task ID: `IMPL-001`
 - `meta.type: "test-gen"`
 - `meta.agent: "@code-developer"`
-- `context.requirements`: Understand source implementation and generate tests
+- `context.requirements`: Understand source implementation and generate tests across all layers (L0-L3)
 - `flow_control.target_files`: Test files to create from TEST_ANALYSIS_RESULTS.md section 5
 
 **Execution Flow**:
 1. **Understand Phase**:
    - Load TEST_ANALYSIS_RESULTS.md and test context
    - Understand source code implementation patterns
-   - Analyze test requirements and conventions
-   - Identify test scenarios and edge cases
+   - Analyze multi-layered test requirements (L0: Static, L1: Unit, L2: Integration, L3: E2E)
+   - Identify test scenarios, edge cases, and error paths
 2. **Generation Phase**:
-   - Generate test files following existing patterns
-   - Ensure test coverage aligns with requirements
+   - Generate L1 unit test files following existing patterns
+   - Generate L2 integration test files (if applicable)
+   - Generate L3 E2E test files (if applicable)
+   - Ensure test coverage aligns with multi-layered requirements
+   - Include both positive and negative test cases
 3. **Verification Phase**:
    - Verify test completeness and correctness
+   - Ensure each test has meaningful assertions
+   - Check for test anti-patterns (tests without assertions, overly broad mocks)
+
+### IMPL-001.5: Test Quality Gate ← **NEW**
+
+**Agent**: `@test-fix-agent`
+
+**Purpose**: Validate test quality before entering fix cycle - prevent "hollow tests" from becoming the source of truth
+
+**Task Configuration**:
+- Task ID: `IMPL-001.5-review`
+- `meta.type: "test-quality-review"`
+- `meta.agent: "@test-fix-agent"`
+- `context.depends_on: ["IMPL-001"]`
+- `context.requirements`: Validate generated tests meet quality standards
+- `context.quality_config`: Load from `.claude/workflows/test-quality-config.json`
+
+**Execution Flow**:
+1. **L0: Static Analysis**:
+   - Run linting on test files (ESLint, Prettier)
+   - Check for test anti-patterns:
+     - Tests without assertions (`expect()` missing)
+     - Empty test bodies (`it('should...', () => {})`)
+     - Disabled tests without justification (`it.skip`, `xit`)
+   - Verify TypeScript type safety (if applicable)
+2. **Coverage Analysis**:
+   - Run coverage analysis on generated tests
+   - Calculate coverage percentage for target source files
+   - Identify uncovered branches and edge cases
+3. **Test Quality Metrics**:
+   - Verify minimum coverage threshold met (default: 80%)
+   - Verify all critical functions have negative test cases
+   - Verify integration tests cover key component interactions
+4. **Quality Gate Decision**:
+   - **PASS**: Coverage ≥ 80%, zero critical anti-patterns → Proceed to IMPL-002
+   - **FAIL**: Coverage < 80% OR critical anti-patterns found → Loop back to IMPL-001 with feedback
+
+**Acceptance Criteria**:
+- Static analysis: Zero critical issues
+- Test coverage: ≥ 80% for target files
+- Test completeness: All targeted functions have unit tests
+- Negative test coverage: Each public API has at least one error handling test
+- Integration coverage: Key component interactions have integration tests (if applicable)
+
+**Failure Handling**:
+If quality gate fails:
+1. Generate detailed feedback report (`.process/test-quality-report.md`)
+2. Update IMPL-001 task with specific improvement requirements
+3. Trigger IMPL-001 re-execution with enhanced context
+4. Maximum 2 quality gate retries before escalating to user
 
 ### IMPL-002: Test Execution & Fix Cycle
 
@@ -350,16 +525,15 @@ Generates minimum 2 tasks (expandable for complex projects):
 - Task ID: `IMPL-002`
 - `meta.type: "test-fix"`
 - `meta.agent: "@test-fix-agent"`
-- `meta.use_codex: true|false` (based on `--use-codex` flag)
 - `context.depends_on: ["IMPL-001"]`
 - `context.requirements`: Execute and fix tests
 
 **Test-Fix Cycle Specification**:
 **Note**: This specification describes what test-cycle-execute orchestrator will do. The agent only executes single tasks.
-- **Cycle Pattern** (orchestrator-managed): test → gemini_diagnose → manual_fix (or codex) → retest
+- **Cycle Pattern** (orchestrator-managed): test → gemini_diagnose → fix (agent or CLI) → retest
 - **Tools Configuration** (orchestrator-controlled):
   - Gemini for analysis with bug-fix template → surgical fix suggestions
-  - Manual fix application (default) OR Codex if `--use-codex` flag (resume mechanism)
+  - Agent fix application (default) OR CLI if `command` field present in implementation_approach
 - **Exit Conditions** (orchestrator-enforced):
   - Success: All tests pass
   - Failure: Max iterations reached (5)
@@ -384,7 +558,7 @@ Generates minimum 2 tasks (expandable for complex projects):
 
 ### Output Files Structure
 
-Created in `.workflow/WFS-test-[session]/`:
+Created in `.workflow/active/WFS-test-[session]/`:
 
 ```
 WFS-test-[session]/
@@ -405,25 +579,69 @@ WFS-test-[session]/
 **File**: `workflow-session.json`
 
 **Session Mode** includes:
-- `workflow_type: "test_session"`
+- `type: "test"` (set by session:start --type test)
 - `source_session_id: "[sourceSessionId]"` (enables automatic cross-session context)
 
 **Prompt Mode** includes:
-- `workflow_type: "test_session"`
+- `type: "test"` (set by session:start --type test)
 - No `source_session_id` field
 
-### Complete Data Flow
+### Execution Flow Diagram
 
-**Example Command**: `/workflow:test-fix-gen WFS-user-auth`
+```
+Test-Fix-Gen Workflow Orchestrator (Dual-Mode Support)
+│
+├─ Phase 1: Create Test Session
+│  ├─ Session Mode: /workflow:session:start --new (with source_session_id)
+│  └─ Prompt Mode: /workflow:session:start --new (without source_session_id)
+│     └─ Returns: testSessionId (WFS-test-[slug])
+│
+├─ Phase 2: Gather Context                                   ← ATTACHED (3 tasks)
+│  ├─ Session Mode: /workflow:tools:test-context-gather
+│  │  └─ Load source session summaries + analyze coverage
+│  └─ Prompt Mode: /workflow:tools:context-gather
+│     └─ Analyze codebase from description
+│     ├─ Phase 2.1: Load context and analyze coverage
+│     ├─ Phase 2.2: Detect test framework and conventions
+│     └─ Phase 2.3: Generate context package
+│     └─ Returns: [test-]context-package.json                ← COLLAPSED
+│
+├─ Phase 3: Test Generation Analysis                         ← ATTACHED (3 tasks)
+│  └─ /workflow:tools:test-concept-enhanced
+│     ├─ Phase 3.1: Analyze coverage gaps with Gemini
+│     ├─ Phase 3.2: Study existing test patterns
+│     └─ Phase 3.3: Generate test generation strategy
+│     └─ Returns: TEST_ANALYSIS_RESULTS.md                   ← COLLAPSED
+│
+├─ Phase 4: Generate Test Tasks                              ← ATTACHED (3 tasks)
+│  └─ /workflow:tools:test-task-generate
+│     ├─ Phase 4.1: Parse TEST_ANALYSIS_RESULTS.md
+│     ├─ Phase 4.2: Generate task JSONs (IMPL-001, IMPL-002)
+│     └─ Phase 4.3: Generate IMPL_PLAN.md and TODO_LIST.md
+│     └─ Returns: Task JSONs and plans                       ← COLLAPSED
+│
+└─ Phase 5: Return Summary
+   └─ Command ends, control returns to user
 
-**Phase Execution Chain**:
-1. Phase 1: `session-start` → `WFS-test-user-auth`
-2. Phase 2: `test-context-gather` → `test-context-package.json`
-3. Phase 3: `test-concept-enhanced` → `TEST_ANALYSIS_RESULTS.md`
-4. Phase 4: `test-task-generate` → `IMPL-001.json` + `IMPL-002.json` (+ additional if needed)
-5. Phase 5: Return summary
+Artifacts Created:
+├── .workflow/active/WFS-test-[session]/
+│   ├── workflow-session.json
+│   ├── IMPL_PLAN.md
+│   ├── TODO_LIST.md
+│   ├── .task/
+│   │   ├── IMPL-001.json (test understanding & generation)
+│   │   ├── IMPL-002.json (test execution & fix cycle)
+│   │   └── IMPL-003.json (optional: test review & certification)
+│   └── .process/
+│       ├── [test-]context-package.json
+│       └── TEST_ANALYSIS_RESULTS.md
 
-**Command completes after Phase 5**
+Key Points:
+• ← ATTACHED: SlashCommand attaches sub-tasks to orchestrator TodoWrite
+• ← COLLAPSED: Sub-tasks executed and collapsed to phase summary
+• Dual-Mode: Session Mode and Prompt Mode share same attachment pattern
+• Command Boundary: Execution delegated to /workflow:test-cycle-execute
+```
 
 ---
 
@@ -459,8 +677,7 @@ WFS-test-[session]/
 4. **Mode Selection**:
    - Use **Session Mode** for completed workflow validation
    - Use **Prompt Mode** for ad-hoc test generation
-   - Use `--use-codex` for autonomous fix application
-   - Use `--cli-execute` for enhanced generation capabilities
+   - Include "use Codex" in description for autonomous fix application
 
 ## Related Commands
 
@@ -473,9 +690,7 @@ WFS-test-[session]/
 - `/workflow:tools:test-context-gather` - Phase 2 (Session Mode): Gather source session context
 - `/workflow:tools:context-gather` - Phase 2 (Prompt Mode): Analyze codebase directly
 - `/workflow:tools:test-concept-enhanced` - Phase 3: Generate test requirements using Gemini
-- `/workflow:tools:test-task-generate` - Phase 4: Generate test task JSONs with fix cycle specification
-- `/workflow:tools:test-task-generate --use-codex` - Phase 4: With automated Codex fixes (when `--use-codex` flag used)
-- `/workflow:tools:test-task-generate --cli-execute` - Phase 4: With CLI execution mode (when `--cli-execute` flag used)
+- `/workflow:tools:test-task-generate` - Phase 4: Generate test task JSONs (CLI tool usage determined semantically)
 
 **Follow-up Commands**:
 - `/workflow:status` - Review generated test tasks

.claude/commands/workflow/test-gen.md

@@ -1,7 +1,7 @@
 ---
 name: test-gen
-description: Create independent test-fix workflow session by analyzing completed implementation
-argument-hint: "[--use-codex] [--cli-execute] source-session-id"
+description: Create independent test-fix workflow session from completed implementation session, analyzes code to generate test tasks
+argument-hint: "source-session-id"
 allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 ---
 
@@ -16,7 +16,20 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 - **Context-First**: Prioritizes gathering code changes and summaries from source session
 - **Format Reuse**: Creates standard `IMPL-*.json` task, using `meta.type: "test-fix"` for agent assignment
 - **Parameter Simplification**: Tools auto-detect test session type via metadata, no manual cross-session parameters needed
-- **Manual First**: Default to manual fixes, use `--use-codex` flag for automated Codex fix application
+- **Semantic CLI Selection**: CLI tool usage is determined by user's task description (e.g., "use Codex for fixes")
+
+**Task Attachment Model**:
+- SlashCommand dispatch **expands workflow** by attaching sub-tasks to current TodoWrite
+- When a sub-command is dispatched (e.g., `/workflow:tools:test-context-gather`), its internal tasks are attached to the orchestrator's TodoWrite
+- Orchestrator **executes these attached tasks** sequentially
+- After completion, attached tasks are **collapsed** back to high-level phase summary
+- This is **task expansion**, not external delegation
+
+**Auto-Continue Mechanism**:
+- TodoList tracks current phase status and dynamically manages task attachment/collapse
+- When each phase finishes executing, automatically execute next pending phase
+- All phases run autonomously without user interaction
+- **⚠️ CONTINUOUS EXECUTION** - Do not stop until all phases complete
 
 **Execution Flow**:
 1. Initialize TodoWrite → Create test session → Parse session ID
@@ -33,23 +46,46 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 3. **Parse Every Output**: Extract required data from each phase for next phase
 4. **Sequential Execution**: Each phase depends on previous phase's output
 5. **Complete All Phases**: Do not return to user until Phase 5 completes (summary returned)
-6. **Track Progress**: Update TodoWrite after every phase completion
+6. **Track Progress**: Update TodoWrite dynamically with task attachment/collapse pattern
 7. **Automatic Detection**: context-gather auto-detects test session and gathers source session context
-8. **Parse --use-codex Flag**: Extract flag from arguments and pass to Phase 4 (test-task-generate)
+8. **Semantic CLI Selection**: CLI tool usage determined from user's task description, passed to Phase 4
 9. **Command Boundary**: This command ends at Phase 5 summary. Test execution is NOT part of this command.
+10. **Task Attachment Model**: SlashCommand dispatch **attaches** sub-tasks to current workflow. Orchestrator **executes** these attached tasks itself, then **collapses** them after completion
+11. **⚠️ CRITICAL: DO NOT STOP**: Continuous multi-phase workflow. After executing all attached tasks, immediately collapse them and execute next phase
 
 ## 5-Phase Execution
 
 ### Phase 1: Create Test Session
-**Command**: `SlashCommand(command="/workflow:session:start --new \"Test validation for [sourceSessionId]\"")`
 
-**Input**: `sourceSessionId` from user argument (e.g., `WFS-user-auth`)
+**Step 1.0: Load Source Session Intent** - Preserve user's original task description for semantic CLI selection
+
+```javascript
+// Read source session metadata to get original task description
+Read(".workflow/active/[sourceSessionId]/workflow-session.json")
+// OR if context-package exists:
+Read(".workflow/active/[sourceSessionId]/.process/context-package.json")
+
+// Extract: metadata.task_description or project/description field
+// This preserves user's CLI tool preferences (e.g., "use Codex for fixes")
+```
+
+**Step 1.1: Dispatch** - Create new test workflow session with preserved intent
+
+```javascript
+// Include original task description to enable semantic CLI selection
+SlashCommand(command="/workflow:session:start --new \"Test validation for [sourceSessionId]: [originalTaskDescription]\"")
+```
+
+**Input**:
+- `sourceSessionId` from user argument (e.g., `WFS-user-auth`)
+- `originalTaskDescription` from source session metadata (preserves CLI tool preferences)
 
 **Expected Behavior**:
 - Creates new session with pattern `WFS-test-[source-slug]` (e.g., `WFS-test-user-auth`)
 - Writes metadata to `workflow-session.json`:
   - `workflow_type: "test_session"`
   - `source_session_id: "[sourceSessionId]"`
+  - Description includes original user intent for semantic CLI selection
 - Returns new session ID for subsequent phases
 
 **Parse Output**:
@@ -67,7 +103,12 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 ---
 
 ### Phase 2: Gather Test Context
-**Command**: `SlashCommand(command="/workflow:tools:test-context-gather --session [testSessionId]")`
+
+**Step 2.1: Dispatch** - Gather test coverage context from source session
+
+```javascript
+SlashCommand(command="/workflow:tools:test-context-gather --session [testSessionId]")
+```
 
 **Input**: `testSessionId` from Phase 1 (e.g., `WFS-test-user-auth`)
 
@@ -89,12 +130,49 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 - Test framework detected
 - Test conventions documented
 
-**TodoWrite**: Mark phase 2 completed, phase 3 in_progress
+<!-- TodoWrite: When test-context-gather dispatched, INSERT 3 test-context-gather tasks -->
+
+**TodoWrite Update (Phase 2 SlashCommand dispatched - tasks attached)**:
+```json
+[
+  {"content": "Create independent test session", "status": "completed", "activeForm": "Creating test session"},
+  {"content": "Phase 2.1: Load source session summaries (test-context-gather)", "status": "in_progress", "activeForm": "Loading source session summaries"},
+  {"content": "Phase 2.2: Analyze test coverage with MCP tools (test-context-gather)", "status": "pending", "activeForm": "Analyzing test coverage"},
+  {"content": "Phase 2.3: Identify coverage gaps and framework (test-context-gather)", "status": "pending", "activeForm": "Identifying coverage gaps"},
+  {"content": "Analyze test requirements with Gemini", "status": "pending", "activeForm": "Analyzing test requirements"},
+  {"content": "Generate test generation and execution tasks", "status": "pending", "activeForm": "Generating test tasks"},
+  {"content": "Return workflow summary", "status": "pending", "activeForm": "Returning workflow summary"}
+]
+```
+
+**Note**: SlashCommand dispatch **attaches** test-context-gather's 3 tasks. Orchestrator **executes** these tasks.
+
+**Next Action**: Tasks attached → **Execute Phase 2.1-2.3** sequentially
+
+<!-- TodoWrite: After Phase 2 tasks complete, REMOVE Phase 2.1-2.3, restore to orchestrator view -->
+
+**TodoWrite Update (Phase 2 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Create independent test session", "status": "completed", "activeForm": "Creating test session"},
+  {"content": "Gather test coverage context", "status": "completed", "activeForm": "Gathering test coverage context"},
+  {"content": "Analyze test requirements with Gemini", "status": "pending", "activeForm": "Analyzing test requirements"},
+  {"content": "Generate test generation and execution tasks", "status": "pending", "activeForm": "Generating test tasks"},
+  {"content": "Return workflow summary", "status": "pending", "activeForm": "Returning workflow summary"}
+]
+```
+
+**Note**: Phase 2 tasks completed and collapsed to summary.
 
 ---
 
 ### Phase 3: Test Generation Analysis
-**Command**: `SlashCommand(command="/workflow:tools:test-concept-enhanced --session [testSessionId] --context [testContextPath]")`
+
+**Step 3.1: Dispatch** - Analyze test requirements with Gemini
+
+```javascript
+SlashCommand(command="/workflow:tools:test-concept-enhanced --session [testSessionId] --context [testContextPath]")
+```
 
 **Input**:
 - `testSessionId` from Phase 1
@@ -121,17 +199,54 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
   - Implementation Targets (test files to create)
   - Success Criteria
 
-**TodoWrite**: Mark phase 3 completed, phase 4 in_progress
+<!-- TodoWrite: When test-concept-enhanced dispatched, INSERT 3 concept-enhanced tasks -->
+
+**TodoWrite Update (Phase 3 SlashCommand dispatched - tasks attached)**:
+```json
+[
+  {"content": "Create independent test session", "status": "completed", "activeForm": "Creating test session"},
+  {"content": "Gather test coverage context", "status": "completed", "activeForm": "Gathering test coverage context"},
+  {"content": "Phase 3.1: Analyze coverage gaps with Gemini (test-concept-enhanced)", "status": "in_progress", "activeForm": "Analyzing coverage gaps"},
+  {"content": "Phase 3.2: Study existing test patterns (test-concept-enhanced)", "status": "pending", "activeForm": "Studying test patterns"},
+  {"content": "Phase 3.3: Generate test generation strategy (test-concept-enhanced)", "status": "pending", "activeForm": "Generating test strategy"},
+  {"content": "Generate test generation and execution tasks", "status": "pending", "activeForm": "Generating test tasks"},
+  {"content": "Return workflow summary", "status": "pending", "activeForm": "Returning workflow summary"}
+]
+```
+
+**Note**: SlashCommand dispatch **attaches** test-concept-enhanced's 3 tasks. Orchestrator **executes** these tasks.
+
+**Next Action**: Tasks attached → **Execute Phase 3.1-3.3** sequentially
+
+<!-- TodoWrite: After Phase 3 tasks complete, REMOVE Phase 3.1-3.3, restore to orchestrator view -->
+
+**TodoWrite Update (Phase 3 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Create independent test session", "status": "completed", "activeForm": "Creating test session"},
+  {"content": "Gather test coverage context", "status": "completed", "activeForm": "Gathering test coverage context"},
+  {"content": "Analyze test requirements with Gemini", "status": "completed", "activeForm": "Analyzing test requirements"},
+  {"content": "Generate test generation and execution tasks", "status": "pending", "activeForm": "Generating test tasks"},
+  {"content": "Return workflow summary", "status": "pending", "activeForm": "Returning workflow summary"}
+]
+```
+
+**Note**: Phase 3 tasks completed and collapsed to summary.
 
 ---
 
 ### Phase 4: Generate Test Tasks
-**Command**: `SlashCommand(command="/workflow:tools:test-task-generate [--use-codex] [--cli-execute] --session [testSessionId]")`
+
+**Step 4.1: Dispatch** - Generate test task JSON files and planning documents
+
+```javascript
+SlashCommand(command="/workflow:tools:test-task-generate --session [testSessionId]")
+```
 
 **Input**:
 - `testSessionId` from Phase 1
-- `--use-codex` flag (if present in original command) - Controls IMPL-002 fix mode
-- `--cli-execute` flag (if present in original command) - Controls IMPL-001 generation mode
+
+**Note**: CLI tool usage for fixes is determined semantically from user's task description (e.g., "use Codex for automated fixes").
 
 **Expected Behavior**:
 - Parse TEST_ANALYSIS_RESULTS.md from Phase 3
@@ -161,19 +276,50 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 - Task ID: `IMPL-002`
 - `meta.type: "test-fix"`
 - `meta.agent: "@test-fix-agent"`
-- `meta.use_codex: true|false` (based on --use-codex flag)
 - `context.depends_on: ["IMPL-001"]`
 - `context.requirements`: Execute and fix tests
 - `flow_control.implementation_approach.test_fix_cycle`: Complete cycle specification
-  - **Cycle pattern**: test → gemini_diagnose → manual_fix (or codex if --use-codex) → retest
-  - **Tools configuration**: Gemini for analysis with bug-fix template, manual or Codex for fixes
+  - **Cycle pattern**: test → gemini_diagnose → fix (agent or CLI based on `command` field) → retest
+  - **Tools configuration**: Gemini for analysis with bug-fix template, agent or CLI for fixes
   - **Exit conditions**: Success (all pass) or failure (max iterations)
 - `flow_control.implementation_approach.modification_points`: 3-phase execution flow
   - Phase 1: Initial test execution
-  - Phase 2: Iterative Gemini diagnosis + manual/Codex fixes (based on flag)
+  - Phase 2: Iterative Gemini diagnosis + fixes (agent or CLI based on step's `command` field)
   - Phase 3: Final validation and certification
 
-**TodoWrite**: Mark phase 4 completed, phase 5 in_progress
+<!-- TodoWrite: When test-task-generate dispatched, INSERT 3 test-task-generate tasks -->
+
+**TodoWrite Update (Phase 4 SlashCommand dispatched - tasks attached)**:
+```json
+[
+  {"content": "Create independent test session", "status": "completed", "activeForm": "Creating test session"},
+  {"content": "Gather test coverage context", "status": "completed", "activeForm": "Gathering test coverage context"},
+  {"content": "Analyze test requirements with Gemini", "status": "completed", "activeForm": "Analyzing test requirements"},
+  {"content": "Phase 4.1: Parse TEST_ANALYSIS_RESULTS.md (test-task-generate)", "status": "in_progress", "activeForm": "Parsing test analysis"},
+  {"content": "Phase 4.2: Generate IMPL-001.json and IMPL-002.json (test-task-generate)", "status": "pending", "activeForm": "Generating task JSONs"},
+  {"content": "Phase 4.3: Generate IMPL_PLAN.md and TODO_LIST.md (test-task-generate)", "status": "pending", "activeForm": "Generating plan documents"},
+  {"content": "Return workflow summary", "status": "pending", "activeForm": "Returning workflow summary"}
+]
+```
+
+**Note**: SlashCommand dispatch **attaches** test-task-generate's 3 tasks. Orchestrator **executes** these tasks.
+
+**Next Action**: Tasks attached → **Execute Phase 4.1-4.3** sequentially
+
+<!-- TodoWrite: After Phase 4 tasks complete, REMOVE Phase 4.1-4.3, restore to orchestrator view -->
+
+**TodoWrite Update (Phase 4 completed - tasks collapsed)**:
+```json
+[
+  {"content": "Create independent test session", "status": "completed", "activeForm": "Creating test session"},
+  {"content": "Gather test coverage context", "status": "completed", "activeForm": "Gathering test coverage context"},
+  {"content": "Analyze test requirements with Gemini", "status": "completed", "activeForm": "Analyzing test requirements"},
+  {"content": "Generate test generation and execution tasks", "status": "completed", "activeForm": "Generating test tasks"},
+  {"content": "Return workflow summary", "status": "in_progress", "activeForm": "Returning workflow summary"}
+]
+```
+
+**Note**: Phase 4 tasks completed and collapsed to summary.
 
 ---
 
@@ -196,7 +342,7 @@ Artifacts Created:
 
 Test Framework: [detected framework]
 Test Files to Generate: [count]
-Fix Mode: [Manual|Codex Automated] (based on --use-codex flag)
+Fix Mode: [Agent|CLI] (based on `command` field in implementation_approach steps)
 
 Review Generated Artifacts:
 - Test plan: .workflow/[testSessionId]/IMPL_PLAN.md
@@ -214,40 +360,75 @@ Ready for execution. Use appropriate workflow commands to proceed.
 
 ## TodoWrite Pattern
 
-Track progress through 5 phases:
+**Core Concept**: Dynamic task attachment and collapse for test-gen workflow with cross-session context gathering and test generation strategy.
 
-```javascript
-TodoWrite({todos: [
-  {"content": "Create independent test session", "status": "in_progress|completed", "activeForm": "Creating test session"},
-  {"content": "Gather test coverage context", "status": "pending|in_progress|completed", "activeForm": "Gathering test coverage context"},
-  {"content": "Analyze test requirements with Gemini", "status": "pending|in_progress|completed", "activeForm": "Analyzing test requirements"},
-  {"content": "Generate test generation and execution tasks", "status": "pending|in_progress|completed", "activeForm": "Generating test tasks"},
-  {"content": "Return workflow summary", "status": "pending|in_progress|completed", "activeForm": "Returning workflow summary"}
-]})
-```
+### Key Principles
 
-Update status to `in_progress` when starting each phase, mark `completed` when done.
+1. **Task Attachment** (when SlashCommand dispatched):
+   - Sub-command's internal tasks are **attached** to orchestrator's TodoWrite
+   - Example: `/workflow:tools:test-context-gather` attaches 3 sub-tasks (Phase 2.1, 2.2, 2.3)
+   - First attached task marked as `in_progress`, others as `pending`
+   - Orchestrator **executes** these attached tasks sequentially
 
-## Data Flow
+2. **Task Collapse** (after sub-tasks complete):
+   - Remove detailed sub-tasks from TodoWrite
+   - **Collapse** to high-level phase summary
+   - Example: Phase 2.1-2.3 collapse to "Gather test coverage context: completed"
+   - Maintains clean orchestrator-level view
+
+3. **Continuous Execution**:
+   - After collapse, automatically proceed to next pending phase
+   - No user intervention required between phases
+   - TodoWrite dynamically reflects current execution state
+
+**Lifecycle Summary**: Initial pending tasks → Phase dispatched (tasks ATTACHED) → Sub-tasks executed sequentially → Phase completed (tasks COLLAPSED to summary) → Next phase begins → Repeat until all phases complete.
+
+### Test-Gen Specific Features
+
+- **Phase 2**: Cross-session context gathering from source implementation session
+- **Phase 3**: Test requirements analysis with Gemini for generation strategy
+- **Phase 4**: Dual-task generation (IMPL-001 for test generation, IMPL-002 for test execution)
+- **Fix Mode Configuration**: CLI tool usage determined semantically from user's task description
+
+
+
+**Note**: See individual Phase descriptions (Phase 2, 3, 4) for detailed TodoWrite Update examples with full JSON structures.
+
+## Execution Flow Diagram
 
 ```
-┌─────────────────────────────────────────────────────────┐
-│ /workflow:test-gen WFS-user-auth                        │
-├─────────────────────────────────────────────────────────┤
-│ Phase 1: session-start → WFS-test-user-auth            │
-│   ↓                                                     │
-│ Phase 2: test-context-gather → test-context-package.json│
-│   ↓                                                     │
-│ Phase 3: test-concept-enhanced → TEST_ANALYSIS_RESULTS.md│
-│   ↓                                                     │
-│ Phase 4: test-task-generate → IMPL-001.json + IMPL-002.json│
-│   ↓                                                     │
-│ Phase 5: Return summary                                │
-└─────────────────────────────────────────────────────────┘
-         COMMAND ENDS - Control returns to user
+Test-Gen Workflow Orchestrator
+│
+├─ Phase 1: Create Test Session
+│  └─ /workflow:session:start --new
+│     └─ Returns: testSessionId (WFS-test-[source])
+│
+├─ Phase 2: Gather Test Context                           ← ATTACHED (3 tasks)
+│  └─ /workflow:tools:test-context-gather
+│     ├─ Phase 2.1: Load source session summaries
+│     ├─ Phase 2.2: Analyze test coverage with MCP tools
+│     └─ Phase 2.3: Identify coverage gaps and framework
+│     └─ Returns: test-context-package.json               ← COLLAPSED
+│
+├─ Phase 3: Test Generation Analysis                      ← ATTACHED (3 tasks)
+│  └─ /workflow:tools:test-concept-enhanced
+│     ├─ Phase 3.1: Analyze coverage gaps with Gemini
+│     ├─ Phase 3.2: Study existing test patterns
+│     └─ Phase 3.3: Generate test generation strategy
+│     └─ Returns: TEST_ANALYSIS_RESULTS.md                ← COLLAPSED
+│
+├─ Phase 4: Generate Test Tasks                           ← ATTACHED (3 tasks)
+│  └─ /workflow:tools:test-task-generate
+│     ├─ Phase 4.1: Parse TEST_ANALYSIS_RESULTS.md
+│     ├─ Phase 4.2: Generate IMPL-001.json and IMPL-002.json
+│     └─ Phase 4.3: Generate IMPL_PLAN.md and TODO_LIST.md
+│     └─ Returns: Task JSONs and plans                    ← COLLAPSED
+│
+└─ Phase 5: Return Summary
+   └─ Command ends, control returns to user
 
 Artifacts Created:
-├── .workflow/WFS-test-[session]/
+├── .workflow/active/WFS-test-[session]/
 │   ├── workflow-session.json
 │   ├── IMPL_PLAN.md
 │   ├── TODO_LIST.md
@@ -257,6 +438,10 @@ Artifacts Created:
 │   └── .process/
 │       ├── test-context-package.json
 │       └── TEST_ANALYSIS_RESULTS.md
+
+Key Points:
+• ← ATTACHED: SlashCommand attaches sub-tasks to orchestrator TodoWrite
+• ← COLLAPSED: Sub-tasks executed and collapsed to phase summary
 ```
 
 ## Session Metadata
@@ -274,7 +459,7 @@ Generates two task definition files:
   - Agent: @test-fix-agent
   - Dependency: IMPL-001 must complete first
   - Max iterations: 5
-  - Fix mode: Manual or Codex (based on --use-codex flag)
+  - Fix mode: Agent or CLI (based on `command` field in implementation_approach)
 
 See `/workflow:tools:test-task-generate` for complete task JSON schemas.
 
@@ -290,7 +475,7 @@ See `/workflow:tools:test-task-generate` for complete task JSON schemas.
 
 ## Output Files
 
-Created in `.workflow/WFS-test-[session]/`:
+Created in `.workflow/active/WFS-test-[session]/`:
 - `workflow-session.json` - Session metadata
 - `.process/test-context-package.json` - Coverage analysis
 - `.process/TEST_ANALYSIS_RESULTS.md` - Test requirements
@@ -311,11 +496,10 @@ Created in `.workflow/WFS-test-[session]/`:
 **IMPL-002.json Structure**:
 - `meta.type: "test-fix"`
 - `meta.agent: "@test-fix-agent"`
-- `meta.use_codex`: true/false (based on --use-codex flag)
 - `context.depends_on: ["IMPL-001"]`
 - `flow_control.implementation_approach.test_fix_cycle`: Complete cycle specification
   - Gemini diagnosis template
-  - Fix application mode (manual/codex)
+  - Fix application mode (agent or CLI based on `command` field)
   - Max iterations: 5
 - `flow_control.implementation_approach.modification_points`: 3-phase flow
 
@@ -333,13 +517,11 @@ See `/workflow:tools:test-task-generate` for complete JSON schemas.
 **Prerequisite Commands**:
 - `/workflow:plan` or `/workflow:execute` - Complete implementation session that needs test validation
 
-**Called by This Command** (5 phases):
+**Dispatched by This Command** (4 phases):
 - `/workflow:session:start` - Phase 1: Create independent test workflow session
 - `/workflow:tools:test-context-gather` - Phase 2: Analyze test coverage and gather source session context
 - `/workflow:tools:test-concept-enhanced` - Phase 3: Generate test requirements and strategy using Gemini
-- `/workflow:tools:test-task-generate` - Phase 4: Generate test generation and execution task JSONs
-- `/workflow:tools:test-task-generate --use-codex` - Phase 4: With automated Codex fixes (when `--use-codex` flag used)
-- `/workflow:tools:test-task-generate --cli-execute` - Phase 4: With CLI execution mode (when `--cli-execute` flag used)
+- `/workflow:tools:test-task-generate` - Phase 4: Generate test task JSONs (CLI tool usage determined semantically)
 
 **Follow-up Commands**:
 - `/workflow:status` - Review generated test tasks

.claude/commands/workflow/tools/conflict-resolution.md

@@ -1,16 +1,16 @@
 ---
 name: conflict-resolution
-description: Detect and resolve conflicts between plan and existing codebase using CLI-powered analysis
+description: Detect and resolve conflicts between plan and existing codebase using CLI-powered analysis with Gemini/Qwen
 argument-hint: "--session WFS-session-id --context path/to/context-package.json"
 examples:
-  - /workflow:tools:conflict-resolution --session WFS-auth --context .workflow/WFS-auth/.process/context-package.json
-  - /workflow:tools:conflict-resolution --session WFS-payment --context .workflow/WFS-payment/.process/context-package.json
+  - /workflow:tools:conflict-resolution --session WFS-auth --context .workflow/active/WFS-auth/.process/context-package.json
+  - /workflow:tools:conflict-resolution --session WFS-payment --context .workflow/active/WFS-payment/.process/context-package.json
 ---
 
 # Conflict Resolution Command
 
 ## Purpose
-Analyzes conflicts between implementation plans and existing codebase, generating multiple resolution strategies.
+Analyzes conflicts between implementation plans and existing codebase, **including module scenario uniqueness detection**, generating multiple resolution strategies with **iterative clarification until boundaries are clear**.
 
 **Scope**: Detection and strategy generation only - NO code modification or task creation.
 
@@ -21,10 +21,14 @@ Analyzes conflicts between implementation plans and existing codebase, generatin
 | Responsibility | Description |
 |---------------|-------------|
 | **Detect Conflicts** | Analyze plan vs existing code inconsistencies |
+| **Scenario Uniqueness** | **NEW**: Search and compare new modules with existing modules for functional overlaps |
 | **Generate Strategies** | Provide 2-4 resolution options per conflict |
+| **Iterative Clarification** | **NEW**: Ask unlimited questions until scenario boundaries are clear and unique |
+| **Agent Re-analysis** | **NEW**: Dynamically update strategies based on user clarifications |
 | **CLI Analysis** | Use Gemini/Qwen (Claude fallback) |
-| **User Decision** | Present options, never auto-apply |
-| **Single Output** | `CONFLICT_RESOLUTION.md` with findings |
+| **User Decision** | Present options ONE BY ONE, never auto-apply |
+| **Direct Text Output** | Output questions via text directly, NEVER use bash echo/printf |
+| **Structured Data** | JSON output for programmatic processing, NO file generation |
 
 ## Conflict Categories
 
@@ -48,6 +52,48 @@ Analyzes conflicts between implementation plans and existing codebase, generatin
 - Setup conflicts
 - Breaking updates
 
+### 5. Module Scenario Overlap
+- **NEW**: Functional overlap between new and existing modules
+- Scenario boundary ambiguity
+- Duplicate responsibility detection
+- Module merge/split decisions
+- **Requires iterative clarification until uniqueness confirmed**
+
+## Execution Process
+
+```
+Input Parsing:
+   ├─ Parse flags: --session, --context
+   └─ Validation: Both REQUIRED, conflict_risk >= medium
+
+Phase 1: Validation
+   ├─ Step 1: Verify session directory exists
+   ├─ Step 2: Load context-package.json
+   ├─ Step 3: Check conflict_risk (skip if none/low)
+   └─ Step 4: Prepare agent task prompt
+
+Phase 2: CLI-Powered Analysis (Agent)
+   ├─ Execute Gemini analysis (Qwen fallback)
+   ├─ Detect conflicts including ModuleOverlap category
+   └─ Generate 2-4 strategies per conflict with modifications
+
+Phase 3: Iterative User Interaction
+   └─ FOR each conflict (one by one):
+      ├─ Display conflict with overlap_analysis (if ModuleOverlap)
+      ├─ Display strategies (2-4 + custom option)
+      ├─ User selects strategy
+      └─ IF clarification_needed:
+         ├─ Collect answers
+         ├─ Agent re-analysis
+         └─ Loop until uniqueness_confirmed (max 10 rounds)
+
+Phase 4: Apply Modifications
+   ├─ Step 1: Extract modifications from resolved strategies
+   ├─ Step 2: Apply using Edit tool
+   ├─ Step 3: Update context-package.json (mark resolved)
+   └─ Step 4: Output custom conflict summary (if any)
+```
+
 ## Execution Flow
 
 ### Phase 1: Validation
@@ -62,34 +108,54 @@ Analyzes conflicts between implementation plans and existing codebase, generatin
 
 **Agent Delegation**:
 ```javascript
-Task(subagent_type="cli-execution-agent", prompt=`
+Task(subagent_type="cli-execution-agent", run_in_background=false, prompt=`
   ## Context
   - Session: {session_id}
   - Risk: {conflict_risk}
   - Files: {existing_files_list}
 
+  ## Exploration Context (from context-package.exploration_results)
+  - Exploration Count: ${contextPackage.exploration_results?.exploration_count || 0}
+  - Angles Analyzed: ${JSON.stringify(contextPackage.exploration_results?.angles || [])}
+  - Pre-identified Conflict Indicators: ${JSON.stringify(contextPackage.exploration_results?.aggregated_insights?.conflict_indicators || [])}
+  - Critical Files: ${JSON.stringify(contextPackage.exploration_results?.aggregated_insights?.critical_files?.map(f => f.path) || [])}
+  - All Patterns: ${JSON.stringify(contextPackage.exploration_results?.aggregated_insights?.all_patterns || [])}
+  - All Integration Points: ${JSON.stringify(contextPackage.exploration_results?.aggregated_insights?.all_integration_points || [])}
+
   ## Analysis Steps
 
+  ### 0. Load Output Schema (MANDATORY)
+  Execute: cat ~/.claude/workflows/cli-templates/schemas/conflict-resolution-schema.json
+
   ### 1. Load Context
   - Read existing files from conflict_detection.existing_files
-  - Load plan from .workflow/{session_id}/.process/context-package.json
+  - Load plan from .workflow/active/{session_id}/.process/context-package.json
+  - **NEW**: Load exploration_results and use aggregated_insights for enhanced analysis
   - Extract role analyses and requirements
 
-  ### 2. Execute CLI Analysis
+  ### 2. Execute CLI Analysis (Enhanced with Exploration + Scenario Uniqueness)
 
   Primary (Gemini):
-  cd {project_root} && gemini -p "
-  PURPOSE: Detect conflicts between plan and codebase
+  ccw cli -p "
+  PURPOSE: Detect conflicts between plan and codebase, using exploration insights
   TASK:
-  • Compare architectures
+  • **Review pre-identified conflict_indicators from exploration results**
+  • Compare architectures (use exploration key_patterns)
   • Identify breaking API changes
   • Detect data model incompatibilities
   • Assess dependency conflicts
+  • **Analyze module scenario uniqueness**
+    - Use exploration integration_points for precise locations
+    - Cross-validate with exploration critical_files
+    - Generate clarification questions for boundary definition
   MODE: analysis
-  CONTEXT: @{existing_files} @.workflow/{session_id}/**/*
-  EXPECTED: Conflict list with severity ratings
-  RULES: Focus on breaking changes and migration needs
-  "
+  CONTEXT: @**/*.ts @**/*.js @**/*.tsx @**/*.jsx @.workflow/active/{session_id}/**/*
+  EXPECTED: Conflict list with severity ratings, including:
+    - Validation of exploration conflict_indicators
+    - ModuleOverlap conflicts with overlap_analysis
+    - Targeted clarification questions
+  RULES: $(cat ~/.claude/workflows/cli-templates/prompts/analysis/02-analyze-code-patterns.txt) | Focus on breaking changes, migration needs, and functional overlaps | Prioritize exploration-identified conflicts | analysis=READ-ONLY
+  " --tool gemini --mode analysis --cd {project_root}
 
   Fallback: Qwen (same prompt) → Claude (manual analysis)
 
@@ -97,261 +163,262 @@ Task(subagent_type="cli-execution-agent", prompt=`
 
   Template per conflict:
   - Severity: Critical/High/Medium
-  - Category: Architecture/API/Data/Dependency
+  - Category: Architecture/API/Data/Dependency/ModuleOverlap
   - Affected files + impact
+  - **For ModuleOverlap**: Include overlap_analysis with existing modules and scenarios
   - Options with pros/cons, effort, risk
+  - **For ModuleOverlap strategies**: Add clarification_needed questions for boundary definition
   - Recommended strategy + rationale
 
   ### 4. Return Structured Conflict Data
 
-  ⚠️ DO NOT generate CONFLICT_RESOLUTION.md file
+  ⚠️ Output to conflict-resolution.json (generated in Phase 4)
 
-  Return JSON format for programmatic processing:
+  **Schema Reference**: Execute \`cat ~/.claude/workflows/cli-templates/schemas/conflict-resolution-schema.json\` to get full schema
 
-  \`\`\`json
-  {
-    "conflicts": [
-      {
-        "id": "CON-001",
-        "brief": "一行中文冲突摘要",
-        "severity": "Critical|High|Medium",
-        "category": "Architecture|API|Data|Dependency",
-        "affected_files": [
-          ".workflow/{session}/.brainstorm/guidance-specification.md",
-          ".workflow/{session}/.brainstorm/system-architect/analysis.md"
-        ],
-        "description": "详细描述冲突 - 什么不兼容",
-        "impact": {
-          "scope": "影响的模块/组件",
-          "compatibility": "Yes|No|Partial",
-          "migration_required": true|false,
-          "estimated_effort": "人天估计"
-        },
-        "strategies": [
-          {
-            "name": "策略名称（中文）",
-            "approach": "实现方法简述",
-            "complexity": "Low|Medium|High",
-            "risk": "Low|Medium|High",
-            "effort": "时间估计",
-            "pros": ["优点1", "优点2"],
-            "cons": ["缺点1", "缺点2"],
-            "modifications": [
-              {
-                "file": ".workflow/{session}/.brainstorm/guidance-specification.md",
-                "section": "## 2. System Architect Decisions",
-                "change_type": "update",
-                "old_content": "原始内容片段（用于定位）",
-                "new_content": "修改后的内容",
-                "rationale": "为什么这样改"
-              },
-              {
-                "file": ".workflow/{session}/.brainstorm/system-architect/analysis.md",
-                "section": "## Design Decisions",
-                "change_type": "update",
-                "old_content": "原始内容片段",
-                "new_content": "修改后的内容",
-                "rationale": "修改理由"
-              }
-            ]
-          },
-          {
-            "name": "策略2名称",
-            "approach": "...",
-            "complexity": "Medium",
-            "risk": "Low",
-            "effort": "1-2天",
-            "pros": ["优点"],
-            "cons": ["缺点"],
-            "modifications": [...]
-          }
-        ],
-        "recommended": 0,
-        "modification_suggestions": [
-          "建议1：具体的修改方向或注意事项",
-          "建议2：可能需要考虑的边界情况",
-          "建议3：相关的最佳实践或模式"
-        ]
-      }
-    ],
-    "summary": {
-      "total": 2,
-      "critical": 1,
-      "high": 1,
-      "medium": 0
-    }
-  }
-  \`\`\`
-
-  ⚠️ CRITICAL Requirements for modifications field:
-  - old_content: Must be exact text from target file (20-100 chars for unique match)
-  - new_content: Complete replacement text (maintains formatting)
-  - change_type: "update" (replace), "add" (insert), "remove" (delete)
-  - file: Full path relative to project root
-  - section: Markdown heading for context (helps locate position)
+  Return JSON following the schema above. Key requirements:
   - Minimum 2 strategies per conflict, max 4
-  - All text in Chinese for user-facing fields (brief, name, pros, cons)
-  - modification_suggestions: 2-5 actionable suggestions for custom handling (Chinese)
-
-  Quality Standards:
-  - Each strategy must have actionable modifications
-  - old_content must be precise enough for Edit tool matching
-  - new_content preserves markdown formatting and structure
-  - Recommended strategy (index) based on lowest complexity + risk
-  - modification_suggestions must be specific, actionable, and context-aware
-  - Each suggestion should address a specific aspect (compatibility, migration, testing, etc.)
+  - All text in Chinese for user-facing fields (brief, name, pros, cons, modification_suggestions)
+  - modifications.old_content: 20-100 chars for unique Edit tool matching
+  - modifications.new_content: preserves markdown formatting
+  - modification_suggestions: 2-5 actionable suggestions for custom handling
 `)
 ```
 
-**Agent Internal Flow**:
+**Agent Internal Flow** (Enhanced):
 ```
 1. Load context package
 2. Check conflict_risk (exit if none/low)
 3. Read existing files + plan artifacts
-4. Run CLI analysis (Gemini→Qwen→Claude)
-5. Parse conflict findings
-6. Generate 2-4 strategies per conflict with modifications
+4. Run CLI analysis (Gemini→Qwen→Claude) with enhanced tasks:
+   - Standard conflict detection (Architecture/API/Data/Dependency)
+   - **NEW: Module scenario uniqueness detection**
+     * Extract new module functionality from plan
+     * Search all existing modules with similar keywords/functionality
+     * Compare scenario coverage and responsibilities
+     * Identify functional overlaps and boundary ambiguities
+     * Generate ModuleOverlap conflicts with overlap_analysis
+5. Parse conflict findings (including ModuleOverlap category)
+6. Generate 2-4 strategies per conflict:
+   - Include modifications for each strategy
+   - **For ModuleOverlap**: Add clarification_needed questions for boundary definition
 7. Return JSON to stdout (NOT file write)
 8. Return execution log path
 ```
 
-### Phase 3: User Confirmation via Text Interaction
-
-**Command parses agent JSON output and presents conflicts to user via text**:
+### Phase 3: User Interaction Loop
 
 ```javascript
-// 1. Parse agent JSON output
-const conflictData = JSON.parse(agentOutput);
-const conflicts = conflictData.conflicts; // No 4-conflict limit
+FOR each conflict:
+  round = 0, clarified = false, userClarifications = []
 
-// 2. Format conflicts as text output (max 10 per round)
-const batchSize = 10;
-const batches = chunkArray(conflicts, batchSize);
+  WHILE (!clarified && round++ < 10):
+    // 1. Display conflict info (text output for context)
+    displayConflictSummary(conflict)  // id, brief, severity, overlap_analysis if ModuleOverlap
 
-for (const [batchIdx, batch] of batches.entries()) {
-  const totalBatches = batches.length;
+    // 2. Strategy selection via AskUserQuestion
+    AskUserQuestion({
+      questions: [{
+        question: formatStrategiesForDisplay(conflict.strategies),
+        header: "策略选择",
+        multiSelect: false,
+        options: [
+          ...conflict.strategies.map((s, i) => ({
+            label: `${s.name}${i === conflict.recommended ? ' (推荐)' : ''}`,
+            description: `${s.complexity}复杂度 | ${s.risk}风险${s.clarification_needed?.length ? ' | ⚠️需澄清' : ''}`
+          })),
+          { label: "自定义修改", description: `建议: ${conflict.modification_suggestions?.slice(0,2).join('; ')}` }
+        ]
+      }]
+    })
 
-  // Output batch header
-  console.log(`===== 冲突解决 (第 ${batchIdx + 1}/${totalBatches} 轮) =====\n`);
-
-  // Output each conflict in batch
-  batch.forEach((conflict, idx) => {
-    const questionNum = batchIdx * batchSize + idx + 1;
-    console.log(`【问题${questionNum} - ${conflict.category}】${conflict.id}: ${conflict.brief}`);
-
-    conflict.strategies.forEach((strategy, sIdx) => {
-      const optionLetter = String.fromCharCode(97 + sIdx); // a, b, c, ...
-      console.log(`${optionLetter}) ${strategy.name}`);
-      console.log(`   说明：${strategy.approach}`);
-      console.log(`   复杂度: ${strategy.complexity} | 风险: ${strategy.risk} | 工作量: ${strategy.effort}`);
-    });
-
-    // Add custom option
-    const customLetter = String.fromCharCode(97 + conflict.strategies.length);
-    console.log(`${customLetter}) 自定义修改`);
-    console.log(`   说明：根据修改建议自行处理，不应用预设策略`);
-
-    // Show modification suggestions
-    if (conflict.modification_suggestions && conflict.modification_suggestions.length > 0) {
-      console.log(`   修改建议：`);
-      conflict.modification_suggestions.forEach(suggestion => {
-        console.log(`   - ${suggestion}`);
-      });
+    // 3. Handle selection
+    if (userChoice === "自定义修改") {
+      customConflicts.push({ id, brief, category, suggestions, overlap_analysis })
+      break
     }
-    console.log();
-  });
 
-  console.log(`请回答 (格式: 1a 2b 3c...)：`);
+    selectedStrategy = findStrategyByName(userChoice)
 
-  // Wait for user input
-  const userInput = await readUserInput();
+    // 4. Clarification (if needed) - batched max 4 per call
+    if (selectedStrategy.clarification_needed?.length > 0) {
+      for (batch of chunk(selectedStrategy.clarification_needed, 4)) {
+        AskUserQuestion({
+          questions: batch.map((q, i) => ({
+            question: q, header: `澄清${i+1}`, multiSelect: false,
+            options: [{ label: "详细说明", description: "提供答案" }]
+          }))
+        })
+        userClarifications.push(...collectAnswers(batch))
+      }
 
-  // Parse answers
-  const answers = parseUserAnswers(userInput, batch);
-}
+      // 5. Agent re-analysis
+      reanalysisResult = Task({
+        subagent_type: "cli-execution-agent",
+        run_in_background: false,
+        prompt: `Conflict: ${conflict.id}, Strategy: ${selectedStrategy.name}
+User Clarifications: ${JSON.stringify(userClarifications)}
+Output: { uniqueness_confirmed, rationale, updated_strategy, remaining_questions }`
+      })
 
-// 3. Build selected strategies (exclude custom selections)
-const selectedStrategies = answers.filter(a => !a.isCustom).map(a => a.strategy);
-const customConflicts = answers.filter(a => a.isCustom).map(a => ({
-  id: a.conflict.id,
-  brief: a.conflict.brief,
-  suggestions: a.conflict.modification_suggestions
-}));
+      if (reanalysisResult.uniqueness_confirmed) {
+        selectedStrategy = { ...reanalysisResult.updated_strategy, clarifications: userClarifications }
+        clarified = true
+      } else {
+        selectedStrategy.clarification_needed = reanalysisResult.remaining_questions
+      }
+    } else {
+      clarified = true
+    }
+
+    if (clarified) resolvedConflicts.push({ conflict, strategy: selectedStrategy })
+  END WHILE
+END FOR
+
+selectedStrategies = resolvedConflicts.map(r => ({
+  conflict_id: r.conflict.id, strategy: r.strategy, clarifications: r.strategy.clarifications || []
+}))
 ```
 
-**Text Output Example**:
-```markdown
-===== 冲突解决 (第 1/1 轮) =====
-
-【问题1 - Architecture】CON-001: 现有认证系统与计划不兼容
-a) 渐进式迁移
-   说明：保留现有系统，逐步迁移到新方案
-   复杂度: Medium | 风险: Low | 工作量: 3-5天
-b) 完全重写
-   说明：废弃旧系统，从零实现新认证
-   复杂度: High | 风险: Medium | 工作量: 7-10天
-c) 自定义修改
-   说明：根据修改建议自行处理，不应用预设策略
-   修改建议：
-   - 评估现有认证系统的兼容性，考虑是否可以通过适配器模式桥接
-   - 检查JWT token格式和验证逻辑是否需要调整
-   - 确保用户会话管理与新架构保持一致
-
-【问题2 - Data】CON-002: 数据库 schema 冲突
-a) 添加迁移脚本
-   说明：创建数据库迁移脚本处理 schema 变更
-   复杂度: Low | 风险: Low | 工作量: 1-2天
-b) 自定义修改
-   说明：根据修改建议自行处理，不应用预设策略
-   修改建议：
-   - 检查现有表结构是否支持新增字段，避免破坏性变更
-   - 考虑使用数据库版本控制工具（如Flyway或Liquibase）
-   - 准备数据迁移和回滚策略
-
-请回答 (格式: 1a 2b)：
-```
-
-**User Input Examples**:
-- `1a 2a` → Conflict 1: 渐进式迁移, Conflict 2: 添加迁移脚本
-- `1b 2b` → Conflict 1: 完全重写, Conflict 2: 自定义修改
-- `1c 2c` → Both choose custom modification (user handles manually with suggestions)
+**Key Points**:
+- AskUserQuestion: max 4 questions/call, batch if more
+- Strategy options: 2-4 strategies + "自定义修改"
+- Clarification loop: max 10 rounds, agent判断 uniqueness_confirmed
+- Custom conflicts: 记录 overlap_analysis 供后续手动处理
 
 ### Phase 4: Apply Modifications
 
 ```javascript
-// 1. Extract modifications from selected strategies
+// 1. Extract modifications from resolved strategies
 const modifications = [];
-selectedStrategies.forEach(strategy => {
-  if (strategy !== "skip") {
-    modifications.push(...strategy.modifications);
+selectedStrategies.forEach(item => {
+  if (item.strategy && item.strategy.modifications) {
+    modifications.push(...item.strategy.modifications.map(mod => ({
+      ...mod,
+      conflict_id: item.conflict_id,
+      clarifications: item.clarifications
+    })));
   }
 });
 
-// 2. Apply each modification using Edit tool
-modifications.forEach(mod => {
-  if (mod.change_type === "update") {
-    Edit({
-      file_path: mod.file,
-      old_string: mod.old_content,
-      new_string: mod.new_content
-    });
+console.log(`\n正在应用 ${modifications.length} 个修改...`);
+
+// 2. Apply each modification using Edit tool (with fallback to context-package.json)
+const appliedModifications = [];
+const failedModifications = [];
+const fallbackConstraints = [];  // For files that don't exist
+
+modifications.forEach((mod, idx) => {
+  try {
+    console.log(`[${idx + 1}/${modifications.length}] 修改 ${mod.file}...`);
+
+    // Check if target file exists (brainstorm files may not exist in lite workflow)
+    if (!file_exists(mod.file)) {
+      console.log(`  ⚠️ 文件不存在，写入 context-package.json 作为约束`);
+      fallbackConstraints.push({
+        source: "conflict-resolution",
+        conflict_id: mod.conflict_id,
+        target_file: mod.file,
+        section: mod.section,
+        change_type: mod.change_type,
+        content: mod.new_content,
+        rationale: mod.rationale
+      });
+      return;  // Skip to next modification
+    }
+
+    if (mod.change_type === "update") {
+      Edit({
+        file_path: mod.file,
+        old_string: mod.old_content,
+        new_string: mod.new_content
+      });
+    } else if (mod.change_type === "add") {
+      // Handle addition - append or insert based on section
+      const fileContent = Read(mod.file);
+      const updated = insertContentAfterSection(fileContent, mod.section, mod.new_content);
+      Write(mod.file, updated);
+    } else if (mod.change_type === "remove") {
+      Edit({
+        file_path: mod.file,
+        old_string: mod.old_content,
+        new_string: ""
+      });
+    }
+
+    appliedModifications.push(mod);
+    console.log(`  ✓ 成功`);
+  } catch (error) {
+    console.log(`  ✗ 失败: ${error.message}`);
+    failedModifications.push({ ...mod, error: error.message });
   }
-  // Handle "add" and "remove" similarly
 });
 
-// 3. Update context-package.json
+// 2b. Generate conflict-resolution.json output file
+const resolutionOutput = {
+  session_id: sessionId,
+  resolved_at: new Date().toISOString(),
+  summary: {
+    total_conflicts: conflicts.length,
+    resolved_with_strategy: selectedStrategies.length,
+    custom_handling: customConflicts.length,
+    fallback_constraints: fallbackConstraints.length
+  },
+  resolved_conflicts: selectedStrategies.map(s => ({
+    conflict_id: s.conflict_id,
+    strategy_name: s.strategy.name,
+    strategy_approach: s.strategy.approach,
+    clarifications: s.clarifications || [],
+    modifications_applied: s.strategy.modifications?.filter(m =>
+      appliedModifications.some(am => am.conflict_id === s.conflict_id)
+    ) || []
+  })),
+  custom_conflicts: customConflicts.map(c => ({
+    id: c.id,
+    brief: c.brief,
+    category: c.category,
+    suggestions: c.suggestions,
+    overlap_analysis: c.overlap_analysis || null
+  })),
+  planning_constraints: fallbackConstraints,  // Constraints for files that don't exist
+  failed_modifications: failedModifications
+};
+
+const resolutionPath = `.workflow/active/${sessionId}/.process/conflict-resolution.json`;
+Write(resolutionPath, JSON.stringify(resolutionOutput, null, 2));
+console.log(`\n📄 冲突解决结果已保存: ${resolutionPath}`);
+
+// 3. Update context-package.json with resolution details (reference to JSON file)
 const contextPackage = JSON.parse(Read(contextPath));
 contextPackage.conflict_detection.conflict_risk = "resolved";
-contextPackage.conflict_detection.resolved_conflicts = conflicts.map(c => c.id);
+contextPackage.conflict_detection.resolution_file = resolutionPath;  // Reference to detailed JSON
+contextPackage.conflict_detection.resolved_conflicts = selectedStrategies.map(s => s.conflict_id);
+contextPackage.conflict_detection.custom_conflicts = customConflicts.map(c => c.id);
 contextPackage.conflict_detection.resolved_at = new Date().toISOString();
 Write(contextPath, JSON.stringify(contextPackage, null, 2));
 
-// 4. Output custom conflict summary (if any)
+// 4. Output custom conflict summary with overlap analysis (if any)
 if (customConflicts.length > 0) {
-  console.log("\n===== 需要自定义处理的冲突 =====\n");
+  console.log(`\n${'='.repeat(60)}`);
+  console.log(`需要自定义处理的冲突 (${customConflicts.length})`);
+  console.log(`${'='.repeat(60)}\n`);
+
   customConflicts.forEach(conflict => {
-    console.log(`【${conflict.id}】${conflict.brief}`);
-    console.log("修改建议：");
+    console.log(`【${conflict.category}】${conflict.id}: ${conflict.brief}`);
+
+    // Show overlap analysis for ModuleOverlap conflicts
+    if (conflict.category === 'ModuleOverlap' && conflict.overlap_analysis) {
+      console.log(`\n场景重叠信息:`);
+      console.log(`  新模块: ${conflict.overlap_analysis.new_module.name}`);
+      console.log(`  场景: ${conflict.overlap_analysis.new_module.scenarios.join(', ')}`);
+      console.log(`\n  与以下模块重叠:`);
+      conflict.overlap_analysis.existing_modules.forEach(mod => {
+        console.log(`    - ${mod.name} (${mod.file})`);
+        console.log(`      重叠场景: ${mod.overlap_scenarios.join(', ')}`);
+      });
+    }
+
+    console.log(`\n修改建议:`);
     conflict.suggestions.forEach(suggestion => {
       console.log(`  - ${suggestion}`);
     });
@@ -359,33 +426,89 @@ if (customConflicts.length > 0) {
   });
 }
 
-// 5. Return summary
+// 5. Output failure summary (if any)
+if (failedModifications.length > 0) {
+  console.log(`\n⚠️ 部分修改失败 (${failedModifications.length}):`);
+  failedModifications.forEach(mod => {
+    console.log(`  - ${mod.file}: ${mod.error}`);
+  });
+}
+
+// 6. Return summary
 return {
-  resolved: modifications.length,
-  custom: customConflicts.length,
-  modified_files: [...new Set(modifications.map(m => m.file))],
-  custom_conflicts: customConflicts
+  total_conflicts: conflicts.length,
+  resolved_with_strategy: selectedStrategies.length,
+  custom_handling: customConflicts.length,
+  modifications_applied: appliedModifications.length,
+  modifications_failed: failedModifications.length,
+  modified_files: [...new Set(appliedModifications.map(m => m.file))],
+  custom_conflicts: customConflicts,
+  clarification_records: selectedStrategies.filter(s => s.clarifications.length > 0)
 };
 ```
 
 **Validation**:
 ```
-✓ Agent returns valid JSON structure
-✓ Text output displays all conflicts (max 10 per round)
-✓ User selections captured correctly
+✓ Agent returns valid JSON structure with ModuleOverlap conflicts
+✓ Conflicts processed ONE BY ONE (not in batches)
+✓ ModuleOverlap conflicts include overlap_analysis field
+✓ Strategies with clarification_needed display questions
+✓ User selections captured correctly per conflict
+✓ Clarification loop continues until uniqueness confirmed
+✓ Agent re-analysis returns uniqueness_confirmed and updated_strategy
+✓ Maximum 10 rounds per conflict safety limit enforced
 ✓ Edit tool successfully applies modifications
 ✓ guidance-specification.md updated
 ✓ Role analyses (*.md) updated
-✓ context-package.json marked as resolved
-✓ Agent log saved to .workflow/{session_id}/.chat/
+✓ context-package.json marked as resolved with clarification records
+✓ Custom conflicts display overlap_analysis for manual handling
+✓ Agent log saved to .workflow/active/{session_id}/.chat/
 ```
 
-## Output Format: Agent JSON Response
+## Output Format
+
+### Primary Output: conflict-resolution.json
+
+**Path**: `.workflow/active/{session_id}/.process/conflict-resolution.json`
+
+**Schema**:
+```json
+{
+  "session_id": "WFS-xxx",
+  "resolved_at": "ISO timestamp",
+  "summary": {
+    "total_conflicts": 3,
+    "resolved_with_strategy": 2,
+    "custom_handling": 1,
+    "fallback_constraints": 0
+  },
+  "resolved_conflicts": [
+    {
+      "conflict_id": "CON-001",
+      "strategy_name": "策略名称",
+      "strategy_approach": "实现方法",
+      "clarifications": [],
+      "modifications_applied": []
+    }
+  ],
+  "custom_conflicts": [
+    {
+      "id": "CON-002",
+      "brief": "冲突摘要",
+      "category": "ModuleOverlap",
+      "suggestions": ["建议1", "建议2"],
+      "overlap_analysis": null
+    }
+  ],
+  "planning_constraints": [],
+  "failed_modifications": []
+}
+```
+
+### Secondary: Agent JSON Response (stdout)
 
 **Focus**: Structured conflict data with actionable modifications for programmatic processing.
 
-**Format**: JSON to stdout (NO file generation)
-
 **Structure**: Defined in Phase 2, Step 4 (agent prompt)
 
 ### Key Requirements
@@ -433,39 +556,47 @@ If Edit tool fails mid-application:
 - Requires: `conflict_risk ≥ medium`
 
 **Output**:
-- Modified files:
-  - `.workflow/{session_id}/.brainstorm/guidance-specification.md`
-  - `.workflow/{session_id}/.brainstorm/{role}/analysis.md`
-  - `.workflow/{session_id}/.process/context-package.json` (conflict_risk → resolved)
-- NO report file generation
+- Generated file:
+  - `.workflow/active/{session_id}/.process/conflict-resolution.json` (primary output)
+- Modified files (if exist):
+  - `.workflow/active/{session_id}/.brainstorm/guidance-specification.md`
+  - `.workflow/active/{session_id}/.brainstorm/{role}/analysis.md`
+  - `.workflow/active/{session_id}/.process/context-package.json` (conflict_risk → resolved, resolution_file reference)
 
 **User Interaction**:
-- Text-based strategy selection (max 10 conflicts per round)
+- **Iterative conflict processing**: One conflict at a time, not in batches
 - Each conflict: 2-4 strategy options + "自定义修改" option (with suggestions)
+- **Clarification loop**: Unlimited questions per conflict until uniqueness confirmed (max 10 rounds)
+- **ModuleOverlap conflicts**: Display overlap_analysis with existing modules
+- **Agent re-analysis**: Dynamic strategy updates based on user clarifications
 
 ### Success Criteria
 ```
-✓ CLI analysis returns valid JSON structure
-✓ Conflicts presented in batches (max 10 per round)
+✓ CLI analysis returns valid JSON structure with ModuleOverlap category
+✓ Agent performs scenario uniqueness detection (searches existing modules)
+✓ Conflicts processed ONE BY ONE with iterative clarification
 ✓ Min 2 strategies per conflict with modifications
+✓ ModuleOverlap conflicts include overlap_analysis with existing modules
+✓ Strategies requiring clarification include clarification_needed questions
 ✓ Each conflict includes 2-5 modification_suggestions
-✓ Text output displays all conflicts correctly with suggestions
-✓ User selections captured and processed
+✓ Text output displays conflict with overlap analysis (if ModuleOverlap)
+✓ User selections captured per conflict
+✓ Clarification loop continues until uniqueness confirmed (unlimited rounds, max 10)
+✓ Agent re-analysis with user clarifications updates strategy
+✓ Uniqueness confirmation based on clear scenario boundaries
 ✓ Edit tool applies modifications successfully
-✓ Custom conflicts displayed with suggestions for manual handling
+✓ Custom conflicts displayed with overlap_analysis for manual handling
 ✓ guidance-specification.md updated with resolved conflicts
 ✓ Role analyses (*.md) updated with resolved conflicts
-✓ context-package.json marked as "resolved"
-✓ No CONFLICT_RESOLUTION.md file generated
-✓ Modification summary includes custom conflict count
-✓ Agent log saved to .workflow/{session_id}/.chat/
+✓ context-package.json marked as "resolved" with clarification records
+✓ conflict-resolution.json generated with full resolution details
+✓ Modification summary includes:
+  - Total conflicts
+  - Resolved with strategy (count)
+  - Custom handling (count)
+  - Clarification records
+  - Overlap analysis for custom ModuleOverlap conflicts
+✓ Agent log saved to .workflow/active/{session_id}/.chat/
 ✓ Error handling robust (validate/retry/degrade)
 ```
 
-## Related Commands
-| Command | Relationship |
-|---------|--------------|
-| `/workflow:tools:context-gather` | Generates input conflict_detection data |
-| `/workflow:plan` | Auto-triggers this when risk ≥ medium |
-| `/workflow:tools:task-generate` | Uses resolved conflicts from updated brainstorm files |
-| `/workflow:brainstorm:artifacts` | Generates guidance-specification.md (modified by this command) |

.claude/commands/workflow/tools/context-gather.md

@@ -1,6 +1,6 @@
 ---
 name: gather
-description: Intelligently collect project context using context-search-agent based on task description and package into standardized JSON
+description: Intelligently collect project context using context-search-agent based on task description, packages into standardized JSON
 argument-hint: "--session WFS-session-id \"task description\""
 examples:
   - /workflow:tools:context-gather --session WFS-user-auth "Implement user authentication system"
@@ -15,14 +15,44 @@ allowed-tools: Task(*), Read(*), Glob(*)
 
 Orchestrator command that invokes `context-search-agent` to gather comprehensive project context for implementation planning. Generates standardized `context-package.json` with codebase analysis, dependencies, and conflict detection.
 
-**Agent**: `context-search-agent` (`.claude/agents/context-search-agent.md`)
 
 ## Core Philosophy
 
 - **Agent Delegation**: Delegate all discovery to `context-search-agent` for autonomous execution
 - **Detection-First**: Check for existing context-package before executing
 - **Plan Mode**: Full comprehensive analysis (vs lightweight brainstorm mode)
-- **Standardized Output**: Generate `.workflow/{session}/.process/context-package.json`
+- **Standardized Output**: Generate `.workflow/active/{session}/.process/context-package.json`
+
+## Execution Process
+
+```
+Input Parsing:
+   ├─ Parse flags: --session
+   └─ Parse: task_description (required)
+
+Step 1: Context-Package Detection
+   └─ Decision (existing package):
+      ├─ Valid package exists → Return existing (skip execution)
+      └─ No valid package → Continue to Step 2
+
+Step 2: Complexity Assessment & Parallel Explore (NEW)
+   ├─ Analyze task_description → classify Low/Medium/High
+   ├─ Select exploration angles (1-4 based on complexity)
+   ├─ Launch N cli-explore-agents in parallel
+   │  └─ Each outputs: exploration-{angle}.json
+   └─ Generate explorations-manifest.json
+
+Step 3: Invoke Context-Search Agent (with exploration input)
+   ├─ Phase 1: Initialization & Pre-Analysis
+   ├─ Phase 2: Multi-Source Discovery
+   │  ├─ Track 0: Exploration Synthesis (prioritize & deduplicate)
+   │  ├─ Track 1-4: Existing tracks
+   └─ Phase 3: Synthesis & Packaging
+      └─ Generate context-package.json with exploration_results
+
+Step 4: Output Verification
+   └─ Verify context-package.json contains exploration_results
+```
 
 ## Execution Flow
 
@@ -48,17 +78,146 @@ if (file_exists(contextPackagePath)) {
 }
 ```
 
-### Step 2: Invoke Context-Search Agent
+### Step 2: Complexity Assessment & Parallel Explore
 
 **Only execute if Step 1 finds no valid package**
 
+```javascript
+// 2.1 Complexity Assessment
+function analyzeTaskComplexity(taskDescription) {
+  const text = taskDescription.toLowerCase();
+  if (/architect|refactor|restructure|modular|cross-module/.test(text)) return 'High';
+  if (/multiple|several|integrate|migrate|extend/.test(text)) return 'Medium';
+  return 'Low';
+}
+
+const ANGLE_PRESETS = {
+  architecture: ['architecture', 'dependencies', 'modularity', 'integration-points'],
+  security: ['security', 'auth-patterns', 'dataflow', 'validation'],
+  performance: ['performance', 'bottlenecks', 'caching', 'data-access'],
+  bugfix: ['error-handling', 'dataflow', 'state-management', 'edge-cases'],
+  feature: ['patterns', 'integration-points', 'testing', 'dependencies'],
+  refactor: ['architecture', 'patterns', 'dependencies', 'testing']
+};
+
+function selectAngles(taskDescription, complexity) {
+  const text = taskDescription.toLowerCase();
+  let preset = 'feature';
+  if (/refactor|architect|restructure/.test(text)) preset = 'architecture';
+  else if (/security|auth|permission/.test(text)) preset = 'security';
+  else if (/performance|slow|optimi/.test(text)) preset = 'performance';
+  else if (/fix|bug|error|issue/.test(text)) preset = 'bugfix';
+
+  const count = complexity === 'High' ? 4 : (complexity === 'Medium' ? 3 : 1);
+  return ANGLE_PRESETS[preset].slice(0, count);
+}
+
+const complexity = analyzeTaskComplexity(task_description);
+const selectedAngles = selectAngles(task_description, complexity);
+const sessionFolder = `.workflow/active/${session_id}/.process`;
+
+// 2.2 Launch Parallel Explore Agents
+const explorationTasks = selectedAngles.map((angle, index) =>
+  Task(
+    subagent_type="cli-explore-agent",
+    run_in_background=false,
+    description=`Explore: ${angle}`,
+    prompt=`
+## Task Objective
+Execute **${angle}** exploration for task planning context. Analyze codebase from this specific angle to discover relevant structure, patterns, and constraints.
+
+## Assigned Context
+- **Exploration Angle**: ${angle}
+- **Task Description**: ${task_description}
+- **Session ID**: ${session_id}
+- **Exploration Index**: ${index + 1} of ${selectedAngles.length}
+- **Output File**: ${sessionFolder}/exploration-${angle}.json
+
+## MANDATORY FIRST STEPS (Execute by Agent)
+**You (cli-explore-agent) MUST execute these steps in order:**
+1. Run: ccw tool exec get_modules_by_depth '{}' (project structure)
+2. Run: rg -l "{keyword_from_task}" --type ts (locate relevant files)
+3. Execute: cat ~/.claude/workflows/cli-templates/schemas/explore-json-schema.json (get output schema reference)
+
+## Exploration Strategy (${angle} focus)
+
+**Step 1: Structural Scan** (Bash)
+- get_modules_by_depth.sh → identify modules related to ${angle}
+- find/rg → locate files relevant to ${angle} aspect
+- Analyze imports/dependencies from ${angle} perspective
+
+**Step 2: Semantic Analysis** (Gemini CLI)
+- How does existing code handle ${angle} concerns?
+- What patterns are used for ${angle}?
+- Where would new code integrate from ${angle} viewpoint?
+
+**Step 3: Write Output**
+- Consolidate ${angle} findings into JSON
+- Identify ${angle}-specific clarification needs
+
+## Expected Output
+
+**File**: ${sessionFolder}/exploration-${angle}.json
+
+**Schema Reference**: Schema obtained in MANDATORY FIRST STEPS step 3, follow schema exactly
+
+**Required Fields** (all ${angle} focused):
+- project_structure: Modules/architecture relevant to ${angle}
+- relevant_files: Files affected from ${angle} perspective
+  **IMPORTANT**: Use object format with relevance scores for synthesis:
+  \`[{path: "src/file.ts", relevance: 0.85, rationale: "Core ${angle} logic"}]\`
+  Scores: 0.7+ high priority, 0.5-0.7 medium, <0.5 low
+- patterns: ${angle}-related patterns to follow
+- dependencies: Dependencies relevant to ${angle}
+- integration_points: Where to integrate from ${angle} viewpoint (include file:line locations)
+- constraints: ${angle}-specific limitations/conventions
+- clarification_needs: ${angle}-related ambiguities (options array + recommended index)
+- _metadata.exploration_angle: "${angle}"
+
+## Success Criteria
+- [ ] Schema obtained via cat explore-json-schema.json
+- [ ] get_modules_by_depth.sh executed
+- [ ] At least 3 relevant files identified with ${angle} rationale
+- [ ] Patterns are actionable (code examples, not generic advice)
+- [ ] Integration points include file:line locations
+- [ ] Constraints are project-specific to ${angle}
+- [ ] JSON output follows schema exactly
+- [ ] clarification_needs includes options + recommended
+
+## Output
+Write: ${sessionFolder}/exploration-${angle}.json
+Return: 2-3 sentence summary of ${angle} findings
+`
+  )
+);
+
+// 2.3 Generate Manifest after all complete
+const explorationFiles = bash(`find ${sessionFolder} -name "exploration-*.json" -type f`).split('\n').filter(f => f.trim());
+const explorationManifest = {
+  session_id,
+  task_description,
+  timestamp: new Date().toISOString(),
+  complexity,
+  exploration_count: selectedAngles.length,
+  angles_explored: selectedAngles,
+  explorations: explorationFiles.map(file => {
+    const data = JSON.parse(Read(file));
+    return { angle: data._metadata.exploration_angle, file: file.split('/').pop(), path: file, index: data._metadata.exploration_index };
+  })
+};
+Write(`${sessionFolder}/explorations-manifest.json`, JSON.stringify(explorationManifest, null, 2));
+```
+
+### Step 3: Invoke Context-Search Agent
+
+**Only execute after Step 2 completes**
+
 ```javascript
 Task(
   subagent_type="context-search-agent",
+  run_in_background=false,
   description="Gather comprehensive context for plan",
   prompt=`
-You are executing as context-search-agent (.claude/agents/context-search-agent.md).
-
 ## Execution Mode
 **PLAN MODE** (Comprehensive) - Full Phase 1-3 execution
 
@@ -67,16 +226,27 @@ You are executing as context-search-agent (.claude/agents/context-search-agent.m
 - **Task Description**: ${task_description}
 - **Output Path**: .workflow/${session_id}/.process/context-package.json
 
+## Exploration Input (from Step 2)
+- **Manifest**: ${sessionFolder}/explorations-manifest.json
+- **Exploration Count**: ${explorationManifest.exploration_count}
+- **Angles**: ${explorationManifest.angles_explored.join(', ')}
+- **Complexity**: ${complexity}
+
 ## Mission
 Execute complete context-search-agent workflow for implementation planning:
 
 ### Phase 1: Initialization & Pre-Analysis
-1. **Detection**: Check for existing context-package (early exit if valid)
-2. **Foundation**: Initialize code-index, get project structure, load docs
-3. **Analysis**: Extract keywords, determine scope, classify complexity
+1. **Project State Loading**:
+   - Read and parse `.workflow/project-tech.json`. Use its `technology_analysis` section as the foundational `project_context`. This is your primary source for architecture, tech stack, and key components.
+   - Read and parse `.workflow/project-guidelines.json`. Load `conventions`, `constraints`, and `learnings` into a `project_guidelines` section.
+   - If files don't exist, proceed with fresh analysis.
+2. **Detection**: Check for existing context-package (early exit if valid)
+3. **Foundation**: Initialize CodexLens, get project structure, load docs
+4. **Analysis**: Extract keywords, determine scope, classify complexity based on task description and project state
 
 ### Phase 2: Multi-Source Context Discovery
-Execute all 4 discovery tracks:
+Execute all discovery tracks:
+- **Track 0**: Exploration Synthesis (load ${sessionFolder}/explorations-manifest.json, prioritize critical_files, deduplicate patterns/integration_points)
 - **Track 1**: Historical archive analysis (query manifest.json for lessons learned)
 - **Track 2**: Reference documentation (CLAUDE.md, architecture docs)
 - **Track 3**: Web examples (use Exa MCP for unfamiliar tech/APIs)
@@ -84,20 +254,24 @@ Execute all 4 discovery tracks:
 
 ### Phase 3: Synthesis, Assessment & Packaging
 1. Apply relevance scoring and build dependency graph
-2. Synthesize 4-source data (archive > docs > code > web)
-3. Integrate brainstorm artifacts (if .brainstorming/ exists, read content)
-4. Perform conflict detection with risk assessment
-5. **Inject historical conflicts** from archive analysis into conflict_detection
-6. Generate and validate context-package.json
+2. **Synthesize 4-source data**: Merge findings from all sources (archive > docs > code > web). **Prioritize the context from `project-tech.json`** for architecture and tech stack unless code analysis reveals it's outdated.
+3. **Populate `project_context`**: Directly use the `technology_analysis` from `project-tech.json` to fill the `project_context` section. Include description, technology_stack, architecture, and key_components.
+4. **Populate `project_guidelines`**: Load conventions, constraints, and learnings from `project-guidelines.json` into a dedicated section.
+5. Integrate brainstorm artifacts (if .brainstorming/ exists, read content)
+6. Perform conflict detection with risk assessment
+7. **Inject historical conflicts** from archive analysis into conflict_detection
+8. Generate and validate context-package.json
 
 ## Output Requirements
 Complete context-package.json with:
 - **metadata**: task_description, keywords, complexity, tech_stack, session_id
-- **project_context**: architecture_patterns, coding_conventions, tech_stack
+- **project_context**: description, technology_stack, architecture, key_components (sourced from `project-tech.json`)
+- **project_guidelines**: {conventions, constraints, quality_rules, learnings} (sourced from `project-guidelines.json`)
 - **assets**: {documentation[], source_code[], config[], tests[]} with relevance scores
 - **dependencies**: {internal[], external[]} with dependency graph
 - **brainstorm_artifacts**: {guidance_specification, role_analyses[], synthesis_output} with content
 - **conflict_detection**: {risk_level, risk_factors, affected_modules[], mitigation_strategy, historical_conflicts[]}
+- **exploration_results**: {manifest_path, exploration_count, angles, explorations[], aggregated_insights} (from Track 0)
 
 ## Quality Validation
 Before completion verify:
@@ -114,7 +288,7 @@ Report completion with statistics.
 )
 ```
 
-### Step 3: Output Verification
+### Step 4: Output Verification
 
 After agent completes, verify output:
 
@@ -124,6 +298,12 @@ const outputPath = `.workflow/${session_id}/.process/context-package.json`;
 if (!file_exists(outputPath)) {
   throw new Error("❌ Agent failed to generate context-package.json");
 }
+
+// Verify exploration_results included
+const pkg = JSON.parse(Read(outputPath));
+if (pkg.exploration_results?.exploration_count > 0) {
+  console.log(`✅ Exploration results aggregated: ${pkg.exploration_results.exploration_count} angles`);
+}
 ```
 
 ## Parameter Reference
@@ -139,11 +319,13 @@ Refer to `context-search-agent.md` Phase 3.7 for complete `context-package.json`
 
 **Key Sections**:
 - **metadata**: Session info, keywords, complexity, tech stack
-- **project_context**: Architecture patterns, conventions, tech stack
+- **project_context**: Architecture patterns, conventions, tech stack (populated from `project-tech.json`)
+- **project_guidelines**: Conventions, constraints, quality rules, learnings (populated from `project-guidelines.json`)
 - **assets**: Categorized files with relevance scores (documentation, source_code, config, tests)
 - **dependencies**: Internal and external dependency graphs
 - **brainstorm_artifacts**: Brainstorm documents with full content (if exists)
 - **conflict_detection**: Risk assessment with mitigation strategies and historical conflicts
+- **exploration_results**: Aggregated exploration insights (from parallel explore phase)
 
 ## Historical Archive Analysis
 
@@ -154,7 +336,7 @@ The context-search-agent MUST perform historical archive analysis as Track 1 in
 **Step 1: Check for Archive Manifest**
 ```bash
 # Check if archive manifest exists
-if [[ -f .workflow/.archives/manifest.json ]]; then
+if [[ -f .workflow/archives/manifest.json ]]; then
   # Manifest available for querying
 fi
 ```
@@ -233,7 +415,7 @@ if (historicalConflicts.length > 0 && currentRisk === "low") {
 ### Archive Query Algorithm
 
 ```markdown
-1. IF .workflow/.archives/manifest.json does NOT exist → Skip Track 1, continue to Track 2
+1. IF .workflow/archives/manifest.json does NOT exist → Skip Track 1, continue to Track 2
 2. IF manifest exists:
    a. Load manifest.json
    b. Extract keywords from task_description (nouns, verbs, technical terms)
@@ -250,33 +432,10 @@ if (historicalConflicts.length > 0 && currentRisk === "low") {
 3. Continue to Track 2 (reference documentation)
 ```
 
-## Usage Examples
-
-### Basic Usage
-```bash
-/workflow:tools:context-gather --session WFS-auth-feature "Implement JWT authentication with refresh tokens"
-```
-## Success Criteria
-
-- ✅ Valid context-package.json generated in `.workflow/{session}/.process/`
-- ✅ Contains >80% relevant files based on task keywords
-- ✅ Execution completes within 2 minutes
-- ✅ All required schema fields present and valid
-- ✅ Conflict risk accurately assessed
-- ✅ Agent reports completion with statistics
-
-## Error Handling
-
-| Error | Cause | Resolution |
-|-------|-------|------------|
-| Package validation failed | Invalid session_id in existing package | Re-run agent to regenerate |
-| Agent execution timeout | Large codebase or slow MCP | Increase timeout, check code-index status |
-| Missing required fields | Agent incomplete execution | Check agent logs, verify schema compliance |
-| File count exceeds limit | Too many relevant files | Agent should auto-prioritize top 50 by relevance |
-
 ## Notes
 
 - **Detection-first**: Always check for existing package before invoking agent
-- **Agent autonomy**: Agent handles all discovery logic per `.claude/agents/context-search-agent.md`
+- **Dual project file integration**: Agent reads both `.workflow/project-tech.json` (tech analysis) and `.workflow/project-guidelines.json` (user constraints) as primary sources
+- **Guidelines injection**: Project guidelines are included in context-package to ensure task generation respects user-defined constraints
 - **No redundancy**: This command is a thin orchestrator, all logic in agent
 - **Plan-specific**: Use this for implementation planning; brainstorm mode uses direct agent call

.claude/commands/workflow/tools/task-generate-agent.md

@@ -1,363 +1,563 @@
 ---
 name: task-generate-agent
-description: Autonomous task generation using action-planning-agent with discovery and output phases
-argument-hint: "--session WFS-session-id [--cli-execute]"
+description: Generate implementation plan documents (IMPL_PLAN.md, task JSONs, TODO_LIST.md) using action-planning-agent - produces planning artifacts, does NOT execute code implementation
+argument-hint: "--session WFS-session-id"
 examples:
   - /workflow:tools:task-generate-agent --session WFS-auth
-  - /workflow:tools:task-generate-agent --session WFS-auth --cli-execute
 ---
 
-# Autonomous Task Generation Command
+# Generate Implementation Plan Command
 
 ## Overview
-Autonomous task JSON and IMPL_PLAN.md generation using action-planning-agent with two-phase execution: discovery and document generation. Supports both agent-driven execution (default) and CLI tool execution modes.
+Generate implementation planning documents (IMPL_PLAN.md, task JSONs, TODO_LIST.md) using action-planning-agent. This command produces **planning artifacts only** - it does NOT execute code implementation. Actual code implementation requires separate execution command (e.g., /workflow:execute).
 
 ## Core Philosophy
-- **Agent-Driven**: Delegate execution to action-planning-agent for autonomous operation
-- **Two-Phase Flow**: Discovery (context gathering) → Output (document generation)
+- **Planning Only**: Generate planning documents (IMPL_PLAN.md, task JSONs, TODO_LIST.md) - does NOT implement code
+- **Agent-Driven Document Generation**: Delegate plan generation to action-planning-agent
+- **N+1 Parallel Planning**: Auto-detect multi-module projects, enable parallel planning (2+1 or 3+1 mode)
+- **Progressive Loading**: Load context incrementally (Core → Selective → On-Demand) due to analysis.md file size
 - **Memory-First**: Reuse loaded documents from conversation memory
+- **Smart Selection**: Load synthesis_output OR guidance + relevant role analyses, NOT all role analyses
 - **MCP-Enhanced**: Use MCP tools for advanced code analysis and research
-- **Pre-Selected Templates**: Command selects correct template based on `--cli-execute` flag **before** invoking agent
-- **Agent Simplicity**: Agent receives pre-selected template and focuses only on content generation
 - **Path Clarity**: All `focus_paths` prefer absolute paths (e.g., `D:\\project\\src\\module`), or clear relative paths from project root (e.g., `./src/module`)
 
-## Execution Lifecycle
+## Execution Process
 
-### Phase 1: Discovery & Context Loading
-**⚡ Memory-First Rule**: Skip file loading if documents already in conversation memory
+```
+Input Parsing:
+   ├─ Parse flags: --session
+   └─ Validation: session_id REQUIRED
 
-**Agent Context Package**:
+Phase 0: User Configuration (Interactive)
+   ├─ Question 1: Supplementary materials/guidelines?
+   ├─ Question 2: Execution method preference (Agent/CLI/Hybrid)
+   ├─ Question 3: CLI tool preference (if CLI selected)
+   └─ Store: userConfig for agent prompt
+
+Phase 1: Context Preparation & Module Detection (Command)
+   ├─ Assemble session paths (metadata, context package, output dirs)
+   ├─ Provide metadata (session_id, execution_mode, mcp_capabilities)
+   ├─ Auto-detect modules from context-package + directory structure
+   └─ Decision:
+      ├─ modules.length == 1 → Single Agent Mode (Phase 2A)
+      └─ modules.length >= 2 → Parallel Mode (Phase 2B + Phase 3)
+
+Phase 2A: Single Agent Planning (Original Flow)
+   ├─ Load context package (progressive loading strategy)
+   ├─ Generate Task JSON Files (.task/IMPL-*.json)
+   ├─ Create IMPL_PLAN.md
+   └─ Generate TODO_LIST.md
+
+Phase 2B: N Parallel Planning (Multi-Module)
+   ├─ Launch N action-planning-agents simultaneously (one per module)
+   ├─ Each agent generates module-scoped tasks (IMPL-{prefix}{seq}.json)
+   ├─ Task ID format: IMPL-A1, IMPL-A2... / IMPL-B1, IMPL-B2...
+   └─ Each module limited to ≤9 tasks
+
+Phase 3: Integration (+1 Coordinator, Multi-Module Only)
+   ├─ Collect all module task JSONs
+   ├─ Resolve cross-module dependencies (CROSS::{module}::{pattern} → actual ID)
+   ├─ Generate unified IMPL_PLAN.md (grouped by module)
+   └─ Generate TODO_LIST.md (hierarchical: module → tasks)
+```
+
+## Document Generation Lifecycle
+
+### Phase 0: User Configuration (Interactive)
+
+**Purpose**: Collect user preferences before task generation to ensure generated tasks match execution expectations.
+
+**User Questions**:
 ```javascript
-{
-  "session_id": "WFS-[session-id]",
-  "execution_mode": "agent-mode" | "cli-execute-mode",  // Determined by flag
-  "task_json_template_path": "~/.claude/workflows/cli-templates/prompts/workflow/task-json-agent-mode.txt"
-                           | "~/.claude/workflows/cli-templates/prompts/workflow/task-json-cli-mode.txt",
-  // Path selected by command based on --cli-execute flag, agent reads it
-  "session_metadata": {
-    // If in memory: use cached content
-    // Else: Load from .workflow/{session-id}/workflow-session.json
-  },
-  "brainstorm_artifacts": {
-    // Loaded from context-package.json → brainstorm_artifacts section
-    "role_analyses": [
-      {
-        "role": "system-architect",
-        "files": [{"path": "...", "type": "primary|supplementary"}]
-      }
-    ],
-    "guidance_specification": {"path": "...", "exists": true},
-    "synthesis_output": {"path": "...", "exists": true},
-    "conflict_resolution": {"path": "...", "exists": true}  // if conflict_risk >= medium
-  },
-  "context_package_path": ".workflow/{session-id}/.process/context-package.json",
-  "context_package": {
-    // If in memory: use cached content
-    // Else: Load from .workflow/{session-id}/.process/context-package.json
-  },
-  "mcp_capabilities": {
-    "code_index": true,
-    "exa_code": true,
-    "exa_web": true
-  }
+AskUserQuestion({
+  questions: [
+    {
+      question: "Do you have supplementary materials or guidelines to include?",
+      header: "Materials",
+      multiSelect: false,
+      options: [
+        { label: "No additional materials", description: "Use existing context only" },
+        { label: "Provide file paths", description: "I'll specify paths to include" },
+        { label: "Provide inline content", description: "I'll paste content directly" }
+      ]
+    },
+    {
+      question: "Select execution method for generated tasks:",
+      header: "Execution",
+      multiSelect: false,
+      options: [
+        { label: "Agent (Recommended)", description: "Claude agent executes tasks directly" },
+        { label: "Hybrid", description: "Agent orchestrates, calls CLI for complex steps" },
+        { label: "CLI Only", description: "All execution via CLI tools (codex/gemini/qwen)" }
+      ]
+    },
+    {
+      question: "If using CLI, which tool do you prefer?",
+      header: "CLI Tool",
+      multiSelect: false,
+      options: [
+        { label: "Codex (Recommended)", description: "Best for implementation tasks" },
+        { label: "Gemini", description: "Best for analysis and large context" },
+        { label: "Qwen", description: "Alternative analysis tool" },
+        { label: "Auto", description: "Let agent decide per-task" }
+      ]
+    }
+  ]
+})
+```
+
+**Handle Materials Response**:
+```javascript
+if (userConfig.materials === "Provide file paths") {
+  // Follow-up question for file paths
+  const pathsResponse = AskUserQuestion({
+    questions: [{
+      question: "Enter file paths to include (comma-separated or one per line):",
+      header: "Paths",
+      multiSelect: false,
+      options: [
+        { label: "Enter paths", description: "Provide paths in text input" }
+      ]
+    }]
+  })
+  userConfig.supplementaryPaths = parseUserPaths(pathsResponse)
 }
 ```
 
-**Discovery Actions**:
-1. **Load Session Context** (if not in memory)
-   ```javascript
-   if (!memory.has("workflow-session.json")) {
-     Read(.workflow/{session-id}/workflow-session.json)
-   }
-   ```
-
-2. **Load Context Package** (if not in memory)
-   ```javascript
-   if (!memory.has("context-package.json")) {
-     Read(.workflow/{session-id}/.process/context-package.json)
-   }
-   ```
-
-3. **Extract & Load Role Analyses** (from context-package.json)
-   ```javascript
-   // Extract role analysis paths from context package
-   const roleAnalysisPaths = contextPackage.brainstorm_artifacts.role_analyses
-     .flatMap(role => role.files.map(f => f.path));
-
-   // Load each role analysis file
-   roleAnalysisPaths.forEach(path => Read(path));
-   ```
-
-4. **Load Conflict Resolution** (from context-package.json, if exists)
-   ```javascript
-   if (contextPackage.brainstorm_artifacts.conflict_resolution?.exists) {
-     Read(contextPackage.brainstorm_artifacts.conflict_resolution.path)
-   }
-   ```
-
-5. **Code Analysis with Native Tools** (optional - enhance understanding)
-   ```bash
-   # Find relevant files for task context
-   find . -name "*auth*" -type f
-   rg "authentication|oauth" -g "*.ts"
-   ```
-
-6. **MCP External Research** (optional - gather best practices)
-   ```javascript
-   // Get external examples for implementation
-   mcp__exa__get_code_context_exa(
-     query="TypeScript JWT authentication best practices",
-     tokensNum="dynamic"
-   )
-   ```
-
-### Phase 2: Agent Execution (Document Generation)
-
-**Pre-Agent Template Selection** (Command decides path before invoking agent):
+**Build userConfig**:
 ```javascript
-// Command checks flag and selects template PATH (not content)
-const templatePath = hasCliExecuteFlag
-  ? "~/.claude/workflows/cli-templates/prompts/workflow/task-json-cli-mode.txt"
-  : "~/.claude/workflows/cli-templates/prompts/workflow/task-json-agent-mode.txt";
+const userConfig = {
+  supplementaryMaterials: {
+    type: "none|paths|inline",
+    content: [...],  // Parsed paths or inline content
+  },
+  executionMethod: "agent|hybrid|cli",
+  preferredCliTool: "codex|gemini|qwen|auto",
+  enableResume: true  // Always enable resume for CLI executions
+}
 ```
 
+**Pass to Agent**: Include `userConfig` in agent prompt for Phase 2A/2B.
+
+### Phase 1: Context Preparation & Module Detection (Command Responsibility)
+
+**Command prepares session paths, metadata, and detects module structure.**
+
+**Session Path Structure**:
+```
+.workflow/active/WFS-{session-id}/
+├── workflow-session.json          # Session metadata
+├── .process/
+│   └── context-package.json       # Context package with artifact catalog
+├── .task/                         # Output: Task JSON files
+│   ├── IMPL-A1.json               # Multi-module: prefixed by module
+│   ├── IMPL-A2.json
+│   ├── IMPL-B1.json
+│   └── ...
+├── IMPL_PLAN.md                   # Output: Implementation plan (grouped by module)
+└── TODO_LIST.md                   # Output: TODO list (hierarchical)
+```
+
+**Command Preparation**:
+1. **Assemble Session Paths** for agent prompt:
+   - `session_metadata_path`
+   - `context_package_path`
+   - Output directory paths
+
+2. **Provide Metadata** (simple values):
+   - `session_id`
+   - `mcp_capabilities` (available MCP tools)
+
+3. **Auto Module Detection** (determines single vs parallel mode):
+   ```javascript
+   function autoDetectModules(contextPackage, projectRoot) {
+     // === Complexity Gate: Only parallelize for High complexity ===
+     const complexity = contextPackage.metadata?.complexity || 'Medium';
+     if (complexity !== 'High') {
+       // Force single agent mode for Low/Medium complexity
+       // This maximizes agent context reuse for related tasks
+       return [{ name: 'main', prefix: '', paths: ['.'] }];
+     }
+
+     // Priority 1: Explicit frontend/backend separation
+     if (exists('src/frontend') && exists('src/backend')) {
+       return [
+         { name: 'frontend', prefix: 'A', paths: ['src/frontend'] },
+         { name: 'backend', prefix: 'B', paths: ['src/backend'] }
+       ];
+     }
+
+     // Priority 2: Monorepo structure
+     if (exists('packages/*') || exists('apps/*')) {
+       return detectMonorepoModules();  // Returns 2-3 main packages
+     }
+
+     // Priority 3: Context-package dependency clustering
+     const modules = clusterByDependencies(contextPackage.dependencies?.internal);
+     if (modules.length >= 2) return modules.slice(0, 3);
+
+     // Default: Single module (original flow)
+     return [{ name: 'main', prefix: '', paths: ['.'] }];
+   }
+   ```
+
+**Decision Logic**:
+- `complexity !== 'High'` → Force Phase 2A (Single Agent, maximize context reuse)
+- `modules.length == 1` → Phase 2A (Single Agent, original flow)
+- `modules.length >= 2 && complexity == 'High'` → Phase 2B + Phase 3 (N+1 Parallel)
+
+**Note**: CLI tool usage is now determined semantically by action-planning-agent based on user's task description, not by flags.
+
+### Phase 2A: Single Agent Planning (Original Flow)
+
+**Condition**: `modules.length == 1` (no multi-module detected)
+
+**Purpose**: Generate IMPL_PLAN.md, task JSONs, and TODO_LIST.md - planning documents only, NOT code implementation.
+
 **Agent Invocation**:
 ```javascript
 Task(
   subagent_type="action-planning-agent",
-  description="Generate task JSON and implementation plan",
+  run_in_background=false,
+  description="Generate planning documents (IMPL_PLAN.md, task JSONs, TODO_LIST.md)",
   prompt=`
-## Execution Context
+## TASK OBJECTIVE
+Generate implementation planning documents (IMPL_PLAN.md, task JSONs, TODO_LIST.md) for workflow session
 
-**Session ID**: WFS-{session-id}
-**Execution Mode**: {agent-mode | cli-execute-mode}
-**Task JSON Template Path**: {template_path}
+IMPORTANT: This is PLANNING ONLY - you are generating planning documents, NOT implementing code.
 
-## Phase 1: Discovery Results (Provided Context)
+CRITICAL: Follow the progressive loading strategy defined in agent specification (load analysis.md files incrementally due to file size)
 
-### Session Metadata
-{session_metadata_content}
+## SESSION PATHS
+Input:
+  - Session Metadata: .workflow/active/{session-id}/workflow-session.json
+  - Context Package: .workflow/active/{session-id}/.process/context-package.json
 
-### Role Analyses (Enhanced by Synthesis)
-{role_analyses_content}
-- Includes requirements, design specs, enhancements, and clarifications from synthesis phase
+Output:
+  - Task Dir: .workflow/active/{session-id}/.task/
+  - IMPL_PLAN: .workflow/active/{session-id}/IMPL_PLAN.md
+  - TODO_LIST: .workflow/active/{session-id}/TODO_LIST.md
 
-### Artifacts Inventory
-- **Guidance Specification**: {guidance_spec_path}
-- **Role Analyses**: {role_analyses_list}
+## CONTEXT METADATA
+Session ID: {session-id}
+MCP Capabilities: {exa_code, exa_web, code_index}
 
-### Context Package
-{context_package_summary}
-- Includes conflict_risk assessment
+## USER CONFIGURATION (from Phase 0)
+Execution Method: ${userConfig.executionMethod}  // agent|hybrid|cli
+Preferred CLI Tool: ${userConfig.preferredCliTool}  // codex|gemini|qwen|auto
+Supplementary Materials: ${userConfig.supplementaryMaterials}
 
-### Conflict Resolution (Conditional)
-If conflict_risk was medium/high, modifications have been applied to:
-- **guidance-specification.md**: Design decisions updated to resolve conflicts
-- **Role analyses (*.md)**: Recommendations adjusted for compatibility
-- **context-package.json**: Marked as "resolved" with conflict IDs
-- NO separate CONFLICT_RESOLUTION.md file (conflicts resolved in-place)
+## CLI TOOL SELECTION
+Based on userConfig.executionMethod:
+- "agent": No command field in implementation_approach steps
+- "hybrid": Add command field to complex steps only (agent handles simple steps)
+- "cli": Add command field to ALL implementation_approach steps
 
-### MCP Analysis Results (Optional)
-**Code Structure**: {mcp_code_index_results}
-**External Research**: {mcp_exa_research_results}
+CLI Resume Support (MANDATORY for all CLI commands):
+- Use --resume parameter to continue from previous task execution
+- Read previous task's cliExecutionId from session state
+- Format: ccw cli -p "[prompt]" --resume ${previousCliId} --tool ${tool} --mode write
 
-## Phase 2: Document Generation Task
+## EXPLORATION CONTEXT (from context-package.exploration_results)
+- Load exploration_results from context-package.json
+- Use aggregated_insights.critical_files for focus_paths generation
+- Apply aggregated_insights.constraints to acceptance criteria
+- Reference aggregated_insights.all_patterns for implementation approach
+- Use aggregated_insights.all_integration_points for precise modification locations
+- Use conflict_indicators for risk-aware task sequencing
 
-### Task Decomposition Standards
-**Core Principle**: Task Merging Over Decomposition
-- **Merge Rule**: Execute together when possible
-- **Decompose Only When**:
-  - Excessive workload (>2500 lines or >6 files)
-  - Different tech stacks or domains
-  - Sequential dependency blocking
-  - Parallel execution needed
+## CONFLICT RESOLUTION CONTEXT (if exists)
+- Check context-package.conflict_detection.resolution_file for conflict-resolution.json path
+- If exists, load .process/conflict-resolution.json:
+  - Apply planning_constraints as task constraints (for brainstorm-less workflows)
+  - Reference resolved_conflicts for implementation approach alignment
+  - Handle custom_conflicts with explicit task notes
 
-**Task Limits**:
-- **Maximum 10 tasks** (hard limit)
-- **Function-based**: Complete units (logic + UI + tests + config)
-- **Hierarchy**: Flat (≤5) | Two-level (6-10) | Re-scope (>10)
+## EXPECTED DELIVERABLES
+1. Task JSON Files (.task/IMPL-*.json)
+   - 6-field schema (id, title, status, context_package_path, meta, context, flow_control)
+   - Quantified requirements with explicit counts
+   - Artifacts integration from context package
+   - **focus_paths enhanced with exploration critical_files**
+   - Flow control with pre_analysis steps (include exploration integration_points analysis)
+   - **CLI Execution IDs and strategies (MANDATORY)**
 
-### Required Outputs
+2. Implementation Plan (IMPL_PLAN.md)
+   - Context analysis and artifact references
+   - Task breakdown and execution strategy
+   - Complete structure per agent definition
 
-#### 1. Task JSON Files (.task/IMPL-*.json)
-**Location**: .workflow/{session-id}/.task/
-**Template**: Read from the template path provided above
+3. TODO List (TODO_LIST.md)
+   - Hierarchical structure (containers, pending, completed markers)
+   - Links to task JSONs and summaries
+   - Matches task JSON hierarchy
 
-**Task JSON Template Loading**:
-\`\`\`
-Read({template_path})
-\`\`\`
+## CLI EXECUTION ID REQUIREMENTS (MANDATORY)
+Each task JSON MUST include:
+- **cli_execution_id**: Unique ID for CLI execution (format: `{session_id}-{task_id}`)
+- **cli_execution**: Strategy object based on depends_on:
+  - No deps → `{ "strategy": "new" }`
+  - 1 dep (single child) → `{ "strategy": "resume", "resume_from": "parent-cli-id" }`
+  - 1 dep (multiple children) → `{ "strategy": "fork", "resume_from": "parent-cli-id" }`
+  - N deps → `{ "strategy": "merge_fork", "merge_from": ["id1", "id2", ...] }`
 
-**Important**:
-- Read the template from the path provided in context
-- Use the template structure exactly as written
-- Replace placeholder variables ({synthesis_spec_path}, {role_analysis_path}, etc.) with actual session-specific paths
-- Include MCP tool integration in pre_analysis steps
-- Map artifacts based on task domain (UI → ui-designer, Backend → system-architect)
+**CLI Execution Strategy Rules**:
+1. **new**: Task has no dependencies - starts fresh CLI conversation
+2. **resume**: Task has 1 parent AND that parent has only this child - continues same conversation
+3. **fork**: Task has 1 parent BUT parent has multiple children - creates new branch with parent context
+4. **merge_fork**: Task has multiple parents - merges all parent contexts into new conversation
 
-#### 2. IMPL_PLAN.md
-**Location**: .workflow/{session-id}/IMPL_PLAN.md
+**Execution Command Patterns**:
+- new: `ccw cli -p "[prompt]" --tool [tool] --mode write --id [cli_execution_id]`
+- resume: `ccw cli -p "[prompt]" --resume [resume_from] --tool [tool] --mode write`
+- fork: `ccw cli -p "[prompt]" --resume [resume_from] --id [cli_execution_id] --tool [tool] --mode write`
+- merge_fork: `ccw cli -p "[prompt]" --resume [merge_from.join(',')] --id [cli_execution_id] --tool [tool] --mode write`
 
-**IMPL_PLAN Template**:
-\`\`\`
-$(cat ~/.claude/workflows/cli-templates/prompts/workflow/impl-plan-template.txt)
-\`\`\`
+## QUALITY STANDARDS
+Hard Constraints:
+  - Task count <= 18 (hard limit - request re-scope if exceeded)
+  - All requirements quantified (explicit counts and enumerated lists)
+  - Acceptance criteria measurable (include verification commands)
+  - Artifact references mapped from context package
+  - All documents follow agent-defined structure
 
-**Important**:
-- Use the template above for IMPL_PLAN.md generation
-- Replace all {placeholder} variables with actual session-specific values
-- Populate CCW Workflow Context based on actual phase progression
-- Extract content from role analyses and context-package.json
-- List all detected brainstorming artifacts with correct paths (role analyses, guidance-specification.md)
-- Include conflict resolution status if CONFLICT_RESOLUTION.md exists
-
-#### 3. TODO_LIST.md
-**Location**: .workflow/{session-id}/TODO_LIST.md
-**Structure**:
-\`\`\`markdown
-# Tasks: {Session Topic}
-
-## Task Progress
-▸ **IMPL-001**: [Main Task Group] → [📋](./.task/IMPL-001.json)
-  - [ ] **IMPL-001.1**: [Subtask] → [📋](./.task/IMPL-001.1.json)
-  - [ ] **IMPL-001.2**: [Subtask] → [📋](./.task/IMPL-001.2.json)
-
-- [ ] **IMPL-002**: [Simple Task] → [📋](./.task/IMPL-002.json)
-
-## Status Legend
-- \`▸\` = Container task (has subtasks)
-- \`- [ ]\` = Pending leaf task
-- \`- [x]\` = Completed leaf task
-\`\`\`
-
-### Execution Instructions for Agent
-
-**Agent Task**: Generate task JSON files, IMPL_PLAN.md, and TODO_LIST.md based on analysis results
-
-**Note**: The correct task JSON template path has been pre-selected by the command based on the `--cli-execute` flag and is provided in the context as `{template_path}`.
-
-**Step 1: Load Task JSON Template**
-- Read template from the provided path: `Read({template_path})`
-- This template is already the correct one based on execution mode
-
-**Step 2: Extract and Decompose Tasks**
-- Parse role analysis.md files for requirements, design specs, and task recommendations
-- Review synthesis enhancements and clarifications in role analyses
-- Apply conflict resolution strategies (if CONFLICT_RESOLUTION.md exists)
-- Apply task merging rules (merge when possible, decompose only when necessary)
-- Map artifacts to tasks based on domain (UI → ui-designer, Backend → system-architect, Data → data-architect)
-- Ensure task count ≤10
-
-**Step 3: Generate Task JSON Files**
-- Use the template structure from Step 1
-- Create .task/IMPL-*.json files with proper structure
-- Replace all {placeholder} variables with actual session paths
-- Embed artifacts array with brainstorming outputs
-- Include MCP tool integration in pre_analysis steps
-
-**Step 4: Create IMPL_PLAN.md**
-- Use IMPL_PLAN template
-- Populate all sections with session-specific content
-- List artifacts with priorities and usage guidelines
-- Document execution strategy and dependencies
-
-**Step 5: Generate TODO_LIST.md**
-- Create task progress checklist matching generated JSONs
-- Use proper status indicators (▸, [ ], [x])
-- Link to task JSON files
-
-**Step 6: Update Session State**
-- Update workflow-session.json with task count and artifact inventory
-- Mark session ready for execution
-
-### MCP Enhancement Examples
-
-**Code Index Usage**:
-\`\`\`javascript
-// Discover authentication-related files
-bash(find . -name "*auth*" -type f)
-
-// Search for OAuth patterns
-bash(rg "oauth|jwt|authentication" -g "*.{ts,js}")
-
-// Get file summary for key components
-bash(rg "^(class|function|export|interface)" src/auth/index.ts)
-\`\`\`
-
-**Exa Research Usage**:
-\`\`\`javascript
-// Get best practices for task implementation
-mcp__exa__get_code_context_exa(
-  query="TypeScript OAuth2 implementation patterns",
-  tokensNum="dynamic"
-)
-
-// Research specific API usage
-mcp__exa__get_code_context_exa(
-  query="Express.js JWT middleware examples",
-  tokensNum=5000
-)
-\`\`\`
-
-### Quality Validation
-
-Before completion, verify:
-- [ ] All task JSON files created in .task/ directory
-- [ ] Each task JSON has 5 required fields
-- [ ] Artifact references correctly mapped
-- [ ] Flow control includes artifact loading steps
-- [ ] MCP tool integration added where appropriate
-- [ ] IMPL_PLAN.md follows required structure
-- [ ] TODO_LIST.md matches task JSONs
-- [ ] Dependency graph is acyclic
-- [ ] Task count within limits (≤10)
-- [ ] Session state updated
-
-## Output
-
-Generate all three documents and report completion status:
-- Task JSON files created: N files
-- Artifacts integrated: synthesis-spec, guidance-specification, N role analyses
-- MCP enhancements: code-index, exa-research
-- Session ready for execution: /workflow:execute
+## SUCCESS CRITERIA
+- All planning documents generated successfully:
+  - Task JSONs valid and saved to .task/ directory
+  - IMPL_PLAN.md created with complete structure
+  - TODO_LIST.md generated matching task JSONs
+- Return completion status with document count and task breakdown summary
 `
 )
 ```
 
+### Phase 2B: N Parallel Planning (Multi-Module)
 
-### Agent Context Passing
+**Condition**: `modules.length >= 2` (multi-module detected)
 
-**Memory-Aware Context Assembly**:
+**Purpose**: Launch N action-planning-agents simultaneously, one per module, for parallel task JSON generation.
+
+**Note**: Phase 2B agents generate Task JSONs ONLY. IMPL_PLAN.md and TODO_LIST.md are generated by Phase 3 Coordinator.
+
+**Parallel Agent Invocation**:
 ```javascript
-// Assemble context package for agent
-const agentContext = {
-  session_id: "WFS-[id]",
+// Launch N agents in parallel (one per module)
+const planningTasks = modules.map(module =>
+  Task(
+    subagent_type="action-planning-agent",
+    run_in_background=false,
+    description=`Generate ${module.name} module task JSONs`,
+    prompt=`
+## TASK OBJECTIVE
+Generate task JSON files for ${module.name} module within workflow session
 
-  // Use memory if available, else load
-  session_metadata: memory.has("workflow-session.json")
-    ? memory.get("workflow-session.json")
-    : Read(.workflow/WFS-[id]/workflow-session.json),
+IMPORTANT: This is PLANNING ONLY - generate task JSONs, NOT implementing code.
+IMPORTANT: Generate Task JSONs ONLY. IMPL_PLAN.md and TODO_LIST.md by Phase 3 Coordinator.
 
-  context_package_path: ".workflow/WFS-[id]/.process/context-package.json",
+CRITICAL: Follow the progressive loading strategy defined in agent specification (load analysis.md files incrementally due to file size)
 
-  context_package: memory.has("context-package.json")
-    ? memory.get("context-package.json")
-    : Read(".workflow/WFS-[id]/.process/context-package.json"),
+## MODULE SCOPE
+- Module: ${module.name} (${module.type})
+- Focus Paths: ${module.paths.join(', ')}
+- Task ID Prefix: IMPL-${module.prefix}
+- Task Limit: ≤9 tasks (hard limit for this module)
+- Other Modules: ${otherModules.join(', ')} (reference only, do NOT generate tasks for them)
 
-  // Extract brainstorm artifacts from context package
-  brainstorm_artifacts: extractBrainstormArtifacts(context_package),
+## SESSION PATHS
+Input:
+  - Session Metadata: .workflow/active/{session-id}/workflow-session.json
+  - Context Package: .workflow/active/{session-id}/.process/context-package.json
 
-  // Load role analyses using paths from context package
-  role_analyses: brainstorm_artifacts.role_analyses
-    .flatMap(role => role.files)
-    .map(file => Read(file.path)),
+Output:
+  - Task Dir: .workflow/active/{session-id}/.task/
 
-  // Load conflict resolution if exists (from context package)
-  conflict_resolution: brainstorm_artifacts.conflict_resolution?.exists
-    ? Read(brainstorm_artifacts.conflict_resolution.path)
-    : null,
+## CONTEXT METADATA
+Session ID: {session-id}
+MCP Capabilities: {exa_code, exa_web, code_index}
 
-  // Optional MCP enhancements
-  mcp_analysis: executeMcpDiscovery()
+## USER CONFIGURATION (from Phase 0)
+Execution Method: ${userConfig.executionMethod}  // agent|hybrid|cli
+Preferred CLI Tool: ${userConfig.preferredCliTool}  // codex|gemini|qwen|auto
+Supplementary Materials: ${userConfig.supplementaryMaterials}
+
+## CLI TOOL SELECTION
+Based on userConfig.executionMethod:
+- "agent": No command field in implementation_approach steps
+- "hybrid": Add command field to complex steps only (agent handles simple steps)
+- "cli": Add command field to ALL implementation_approach steps
+
+CLI Resume Support (MANDATORY for all CLI commands):
+- Use --resume parameter to continue from previous task execution
+- Read previous task's cliExecutionId from session state
+- Format: ccw cli -p "[prompt]" --resume ${previousCliId} --tool ${tool} --mode write
+
+## EXPLORATION CONTEXT (from context-package.exploration_results)
+- Load exploration_results from context-package.json
+- Filter for ${module.name} module: Use aggregated_insights.critical_files matching ${module.paths.join(', ')}
+- Apply module-relevant constraints from aggregated_insights.constraints
+- Reference aggregated_insights.all_patterns applicable to ${module.name}
+- Use aggregated_insights.all_integration_points for precise modification locations within module scope
+- Use conflict_indicators for risk-aware task sequencing
+
+## CONFLICT RESOLUTION CONTEXT (if exists)
+- Check context-package.conflict_detection.resolution_file for conflict-resolution.json path
+- If exists, load .process/conflict-resolution.json:
+  - Apply planning_constraints relevant to ${module.name} as task constraints
+  - Reference resolved_conflicts affecting ${module.name} for implementation approach alignment
+  - Handle custom_conflicts with explicit task notes
+
+## CROSS-MODULE DEPENDENCIES
+- For dependencies ON other modules: Use placeholder depends_on: ["CROSS::{module}::{pattern}"]
+- Example: depends_on: ["CROSS::B::api-endpoint"] (this module depends on B's api-endpoint task)
+- Phase 3 Coordinator resolves to actual task IDs
+- For dependencies FROM other modules: Document in task context as "provides_for" annotation
+
+## EXPECTED DELIVERABLES
+Task JSON Files (.task/IMPL-${module.prefix}*.json):
+  - 6-field schema (id, title, status, context_package_path, meta, context, flow_control)
+  - Task ID format: IMPL-${module.prefix}1, IMPL-${module.prefix}2, ...
+  - Quantified requirements with explicit counts
+  - Artifacts integration from context package (filtered for ${module.name})
+  - **focus_paths enhanced with exploration critical_files (module-scoped)**
+  - Flow control with pre_analysis steps (include exploration integration_points analysis)
+  - **CLI Execution IDs and strategies (MANDATORY)**
+  - Focus ONLY on ${module.name} module scope
+
+## CLI EXECUTION ID REQUIREMENTS (MANDATORY)
+Each task JSON MUST include:
+- **cli_execution_id**: Unique ID for CLI execution (format: `{session_id}-IMPL-${module.prefix}{seq}`)
+- **cli_execution**: Strategy object based on depends_on:
+  - No deps → `{ "strategy": "new" }`
+  - 1 dep (single child) → `{ "strategy": "resume", "resume_from": "parent-cli-id" }`
+  - 1 dep (multiple children) → `{ "strategy": "fork", "resume_from": "parent-cli-id" }`
+  - N deps → `{ "strategy": "merge_fork", "merge_from": ["id1", "id2", ...] }`
+  - Cross-module dep → `{ "strategy": "cross_module_fork", "resume_from": "CROSS::{module}::{pattern}" }`
+
+**CLI Execution Strategy Rules**:
+1. **new**: Task has no dependencies - starts fresh CLI conversation
+2. **resume**: Task has 1 parent AND that parent has only this child - continues same conversation
+3. **fork**: Task has 1 parent BUT parent has multiple children - creates new branch with parent context
+4. **merge_fork**: Task has multiple parents - merges all parent contexts into new conversation
+5. **cross_module_fork**: Task depends on task from another module - Phase 3 resolves placeholder
+
+**Execution Command Patterns**:
+- new: `ccw cli -p "[prompt]" --tool [tool] --mode write --id [cli_execution_id]`
+- resume: `ccw cli -p "[prompt]" --resume [resume_from] --tool [tool] --mode write`
+- fork: `ccw cli -p "[prompt]" --resume [resume_from] --id [cli_execution_id] --tool [tool] --mode write`
+- merge_fork: `ccw cli -p "[prompt]" --resume [merge_from.join(',')] --id [cli_execution_id] --tool [tool] --mode write`
+- cross_module_fork: (Phase 3 resolves placeholder, then uses fork pattern)
+
+## QUALITY STANDARDS
+Hard Constraints:
+  - Task count <= 9 for this module (hard limit - coordinate with Phase 3 if exceeded)
+  - All requirements quantified (explicit counts and enumerated lists)
+  - Acceptance criteria measurable (include verification commands)
+  - Artifact references mapped from context package (module-scoped filter)
+  - Focus paths use absolute paths or clear relative paths from project root
+  - Cross-module dependencies use CROSS:: placeholder format
+
+## SUCCESS CRITERIA
+- Task JSONs saved to .task/ with IMPL-${module.prefix}* naming
+- All task JSONs include cli_execution_id and cli_execution strategy
+- Cross-module dependencies use CROSS:: placeholder format consistently
+- Focus paths scoped to ${module.paths.join(', ')} only
+- Return: task count, task IDs, dependency summary (internal + cross-module)
+    `
+  )
+);
+
+// Execute all in parallel
+await Promise.all(planningTasks);
+```
+
+**Output Structure** (direct to .task/):
+```
+.task/
+├── IMPL-A1.json      # Module A (e.g., frontend)
+├── IMPL-A2.json
+├── IMPL-B1.json      # Module B (e.g., backend)
+├── IMPL-B2.json
+└── IMPL-C1.json      # Module C (e.g., shared)
+```
+
+**Task ID Naming**:
+- Format: `IMPL-{prefix}{seq}.json`
+- Prefix: A, B, C... (assigned by detection order)
+- Sequence: 1, 2, 3... (per-module increment)
+
+### Phase 3: Integration (+1 Coordinator Agent, Multi-Module Only)
+
+**Condition**: Only executed when `modules.length >= 2`
+
+**Purpose**: Collect all module tasks, resolve cross-module dependencies, generate unified IMPL_PLAN.md and TODO_LIST.md documents.
+
+**Coordinator Agent Invocation**:
+```javascript
+// Wait for all Phase 2B agents to complete
+const moduleResults = await Promise.all(planningTasks);
+
+// Launch +1 Coordinator Agent
+Task(
+  subagent_type="action-planning-agent",
+  run_in_background=false,
+  description="Integrate module tasks and generate unified documents",
+  prompt=`
+## TASK OBJECTIVE
+Integrate all module task JSONs, resolve cross-module dependencies, and generate unified IMPL_PLAN.md and TODO_LIST.md
+
+IMPORTANT: This is INTEGRATION ONLY - consolidate existing task JSONs, NOT creating new tasks.
+
+## SESSION PATHS
+Input:
+  - Session Metadata: .workflow/active/{session-id}/workflow-session.json
+  - Context Package: .workflow/active/{session-id}/.process/context-package.json
+  - Task JSONs: .workflow/active/{session-id}/.task/IMPL-*.json (from Phase 2B)
+Output:
+  - Updated Task JSONs: .workflow/active/{session-id}/.task/IMPL-*.json (resolved dependencies)
+  - IMPL_PLAN: .workflow/active/{session-id}/IMPL_PLAN.md
+  - TODO_LIST: .workflow/active/{session-id}/TODO_LIST.md
+
+## CONTEXT METADATA
+Session ID: {session-id}
+Modules: ${modules.map(m => m.name + '(' + m.prefix + ')').join(', ')}
+Module Count: ${modules.length}
+
+## INTEGRATION STEPS
+1. Collect all .task/IMPL-*.json, group by module prefix
+2. Resolve CROSS:: dependencies → actual task IDs, update task JSONs
+3. Generate IMPL_PLAN.md (multi-module format per agent specification)
+4. Generate TODO_LIST.md (hierarchical format per agent specification)
+
+## CROSS-MODULE DEPENDENCY RESOLUTION
+- Pattern: CROSS::{module}::{pattern} → IMPL-{module}* matching title/context
+- Example: CROSS::B::api-endpoint → IMPL-B1 (if B1 title contains "api-endpoint")
+- Log unresolved as warnings
+
+## EXPECTED DELIVERABLES
+1. Updated Task JSONs with resolved dependency IDs
+2. IMPL_PLAN.md - multi-module format with cross-dependency section
+3. TODO_LIST.md - hierarchical by module with cross-dependency section
+
+## SUCCESS CRITERIA
+- No CROSS:: placeholders remaining in task JSONs
+- IMPL_PLAN.md and TODO_LIST.md generated with multi-module structure
+- Return: task count, per-module breakdown, resolved dependency count
+  `
+)
+```
+
+**Dependency Resolution Algorithm**:
+```javascript
+function resolveCrossModuleDependency(placeholder, allTasks) {
+  const [, targetModule, pattern] = placeholder.match(/CROSS::(\w+)::(.+)/);
+  const candidates = allTasks.filter(t =>
+    t.id.startsWith(`IMPL-${targetModule}`) &&
+    (t.title.toLowerCase().includes(pattern.toLowerCase()) ||
+     t.context?.description?.toLowerCase().includes(pattern.toLowerCase()))
+  );
+  return candidates.length > 0
+    ? candidates.sort((a, b) => a.id.localeCompare(b.id))[0].id
+    : placeholder; // Keep for manual resolution
 }
 ```
+

.claude/commands/workflow/tools/task-generate-tdd.md

@@ -1,14 +1,27 @@
 ---
 name: task-generate-tdd
-description: Generate TDD task chains with Red-Green-Refactor dependencies
-argument-hint: "--session WFS-session-id [--agent]"
-allowed-tools: Read(*), Write(*), Bash(gemini:*), TodoWrite(*)
+description: Autonomous TDD task generation using action-planning-agent with Red-Green-Refactor cycles, test-first structure, and cycle validation
+argument-hint: "--session WFS-session-id"
+examples:
+  - /workflow:tools:task-generate-tdd --session WFS-auth
 ---
 
-# TDD Task Generation Command
+# Autonomous TDD Task Generation Command
 
 ## Overview
-Generate TDD-specific tasks from analysis results with complete Red-Green-Refactor cycles contained within each task.
+Autonomous TDD task JSON and IMPL_PLAN.md generation using action-planning-agent with two-phase execution: discovery and document generation. Generates complete Red-Green-Refactor cycles contained within each task.
+
+## Core Philosophy
+- **Agent-Driven**: Delegate execution to action-planning-agent for autonomous operation
+- **Two-Phase Flow**: Discovery (context gathering) → Output (document generation)
+- **Memory-First**: Reuse loaded documents from conversation memory
+- **MCP-Enhanced**: Use MCP tools for advanced code analysis and research
+- **Semantic CLI Selection**: CLI tool usage determined from user's task description, not flags
+- **Agent Simplicity**: Agent generates content with semantic CLI detection
+- **Path Clarity**: All `focus_paths` prefer absolute paths (e.g., `D:\\project\\src\\module`), or clear relative paths from project root (e.g., `./src/module`)
+- **TDD-First**: Every feature starts with a failing test (Red phase)
+- **Feature-Complete Tasks**: Each task contains complete Red-Green-Refactor cycle
+- **Quantification-Enforced**: All test cases, coverage requirements, and implementation scope MUST include explicit counts and enumerations
 
 ## Task Strategy & Philosophy
 
@@ -16,11 +29,6 @@ Generate TDD-specific tasks from analysis results with complete Red-Green-Refact
 - **1 feature = 1 task** containing complete TDD cycle internally
 - Each task executes Red-Green-Refactor phases sequentially
 - Task count = Feature count (typically 5 features = 5 tasks)
-- **Benefits**:
-  - 70% reduction in task management overhead
-  - Continuous context per feature (no switching between TEST/IMPL/REFACTOR)
-  - Simpler dependency management
-  - Maintains TDD rigor through internal phase structure
 
 **Previous Approach** (Deprecated):
 - 1 feature = 3 separate tasks (TEST-N.M, IMPL-N.M, REFACTOR-N.M)
@@ -34,373 +42,361 @@ Generate TDD-specific tasks from analysis results with complete Red-Green-Refact
 - Different tech stacks or domains within feature
 
 ### Task Limits
-- **Maximum 10 tasks** (hard limit for TDD workflows)
+- **Maximum 18 tasks** (hard limit for TDD workflows)
 - **Feature-based**: Complete functional units with internal TDD cycles
 - **Hierarchy**: Flat (≤5 simple features) | Two-level (6-10 for complex features with sub-features)
-- **Re-scope**: If >10 tasks needed, break project into multiple TDD workflow sessions
+- **Re-scope**: If >18 tasks needed, break project into multiple TDD workflow sessions
 
 ### TDD Cycle Mapping
 - **Old approach**: 1 feature = 3 tasks (TEST-N.M, IMPL-N.M, REFACTOR-N.M)
 - **Current approach**: 1 feature = 1 task (IMPL-N with internal Red-Green-Refactor phases)
 - **Complex features**: 1 container (IMPL-N) + subtasks (IMPL-N.M) when necessary
 
-### Core Principles
-- **TDD-First**: Every feature starts with a failing test (Red phase)
-- **Feature-Complete Tasks**: Each task contains complete Red-Green-Refactor cycle
-- **Phase-Explicit**: Internal phases clearly marked in flow_control.implementation_approach
-- **Task Merging**: Prefer single task per feature over decomposition
-- **Path Clarity**: All `focus_paths` prefer absolute paths (e.g., `D:\\project\\src\\module`), or clear relative paths from project root (e.g., `./src/module`)
-- **Artifact-Aware**: Integrates brainstorming outputs
-- **Memory-First**: Reuse loaded documents from memory
-- **Context-Aware**: Analyzes existing codebase and test patterns
-- **Iterative Green Phase**: Auto-diagnose and fix test failures with Gemini + optional Codex
-- **Safety-First**: Auto-revert on max iterations to prevent broken state
+## Execution Process
 
-## Core Responsibilities
-- Parse analysis results and identify testable features
-- Generate feature-complete tasks with internal TDD cycles (1 task per simple feature)
-- Apply task merging strategy by default, create subtasks only when complexity requires
-- Generate IMPL_PLAN.md with TDD Implementation Tasks section
-- Generate TODO_LIST.md with internal TDD phase indicators
-- Update session state for TDD execution with task count compliance
+```
+Input Parsing:
+   ├─ Parse flags: --session
+   └─ Validation: session_id REQUIRED
+
+Phase 1: Discovery & Context Loading (Memory-First)
+   ├─ Load session context (if not in memory)
+   ├─ Load context package (if not in memory)
+   ├─ Load test context package (if not in memory)
+   ├─ Extract & load role analyses from context package
+   ├─ Load conflict resolution (if exists)
+   └─ Optional: MCP external research
+
+Phase 2: Agent Execution (Document Generation)
+   ├─ Pre-agent template selection (semantic CLI detection)
+   ├─ Invoke action-planning-agent
+   ├─ Generate TDD Task JSON Files (.task/IMPL-*.json)
+   │  └─ Each task: complete Red-Green-Refactor cycle internally
+   ├─ Create IMPL_PLAN.md (TDD variant)
+   └─ Generate TODO_LIST.md with TDD phase indicators
+```
 
 ## Execution Lifecycle
 
-### Phase 1: Input Validation & Discovery
-**Memory-First Rule**: Skip file loading if documents already in conversation memory
+### Phase 1: Discovery & Context Loading
+**⚡ Memory-First Rule**: Skip file loading if documents already in conversation memory
 
-1. **Session Validation**
-   - If session metadata in memory → Skip loading
-   - Else: Load `.workflow/{session_id}/workflow-session.json`
-
-2. **Conflict Resolution Check** (NEW - Priority Input)
-   - If CONFLICT_RESOLUTION.md exists → Load selected strategies
-   - Else: Skip to brainstorming artifacts
-   - Path: `.workflow/{session_id}/.process/CONFLICT_RESOLUTION.md`
-
-3. **Artifact Discovery**
-   - If artifact inventory in memory → Skip scanning
-   - Else: Scan `.workflow/{session_id}/.brainstorming/` directory
-   - Detect: role analysis documents, guidance-specification.md, role analyses
-
-4. **Context Package Loading**
-   - Load `.workflow/{session_id}/.process/context-package.json`
-   - Load `.workflow/{session_id}/.process/test-context-package.json` (if exists)
-
-### Phase 2: TDD Task JSON Generation
-
-**Input Sources** (priority order):
-1. **Conflict Resolution** (if exists): `.process/CONFLICT_RESOLUTION.md` - Selected resolution strategies
-2. **Brainstorming Artifacts**: Role analysis documents (system-architect, product-owner, etc.)
-3. **Context Package**: `.process/context-package.json` - Project structure and requirements
-4. **Test Context**: `.process/test-context-package.json` - Existing test patterns
-
-**TDD Task Structure includes**:
-- Feature list with testable requirements
-- Test cases for Red phase
-- Implementation requirements for Green phase (with test-fix cycle)
-- Refactoring opportunities
-- Task dependencies and execution order
-- Conflict resolution decisions (if applicable)
-
-### Phase 3: Task JSON & IMPL_PLAN.md Generation
-
-#### Task Structure (Feature-Complete with Internal TDD)
-For each feature, generate task(s) with ID format:
-- **IMPL-N** - Single task containing complete TDD cycle (Red-Green-Refactor)
-- **IMPL-N.M** - Sub-tasks only when feature is complex (>2500 lines or technical blocking)
-
-**Task Dependency Rules**:
-- **Sequential features**: IMPL-2 depends_on ["IMPL-1"] if Feature 2 needs Feature 1
-- **Independent features**: No dependencies, can execute in parallel
-- **Complex features**: IMPL-N.2 depends_on ["IMPL-N.1"] for subtask ordering
-
-**Agent Assignment**:
-- **All IMPL tasks** → `@code-developer` (handles full TDD cycle)
-- Agent executes Red, Green, Refactor phases sequentially within task
-
-**Meta Fields**:
-- `meta.type`: "feature" (TDD-driven feature implementation)
-- `meta.agent`: "@code-developer"
-- `meta.tdd_workflow`: true (enables TDD-specific flow)
-- `meta.tdd_phase`: Not used (phases are in flow_control.implementation_approach)
-- `meta.max_iterations`: 3 (for Green phase test-fix cycle)
-- `meta.use_codex`: false (manual fixes by default)
-
-#### Task JSON Structure Reference
-
-**Simple Feature Task (IMPL-N.json)** - Recommended for most features:
-```json
+**Agent Context Package**:
+```javascript
 {
-  "id": "IMPL-N",                                  // Task identifier
-  "title": "Feature description with TDD",         // Human-readable title
-  "status": "pending",                             // pending | in_progress | completed | container
-  "context_package_path": ".workflow/{session-id}/.process/context-package.json", // Path to smart context package
-  "meta": {
-    "type": "feature",                             // Task type
-    "agent": "@code-developer",                    // Assigned agent
-    "tdd_workflow": true,                          // REQUIRED: Enables TDD flow
-    "max_iterations": 3,                           // Green phase test-fix cycle limit
-    "use_codex": false                             // false=manual fixes, true=Codex automated fixes
+  "session_id": "WFS-[session-id]",
+  "workflow_type": "tdd",
+  // Note: CLI tool usage is determined semantically by action-planning-agent based on user's task description
+  "session_metadata": {
+    // If in memory: use cached content
+    // Else: Load from .workflow/active//{session-id}/workflow-session.json
   },
-  "context": {
-    "requirements": [                              // Feature requirements with TDD phases
-      "Feature description",
-      "Red: Test scenarios to write",
-      "Green: Implementation approach with test-fix cycle",
-      "Refactor: Code quality improvements"
-    ],
-    "tdd_cycles": [                                // OPTIONAL: Detailed test cycles
+  "brainstorm_artifacts": {
+    // Loaded from context-package.json → brainstorm_artifacts section
+    "role_analyses": [
       {
-        "cycle": 1,
-        "feature": "Specific functionality",
-        "test_focus": "What to test",
-        "expected_failure": "Why test should fail initially"
+        "role": "system-architect",
+        "files": [{"path": "...", "type": "primary|supplementary"}]
       }
     ],
-    "focus_paths": ["D:\\project\\src\\path", "./tests/path"],  // Absolute or clear relative paths from project root
-    "acceptance": [                                // Success criteria
-      "All tests pass (Red → Green)",
-      "Code refactored (Refactor complete)",
-      "Test coverage ≥80%"
-    ],
-    "depends_on": []                               // Task dependencies
+    "guidance_specification": {"path": "...", "exists": true},
+    "synthesis_output": {"path": "...", "exists": true},
+    "conflict_resolution": {"path": "...", "exists": true}  // if conflict_risk >= medium
   },
-  "flow_control": {
-    "pre_analysis": [                              // OPTIONAL: Pre-execution checks
-      {
-        "step": "check_test_framework",
-        "action": "Verify test framework",
-        "command": "bash(npm list jest)",
-        "output_to": "test_framework_info",
-        "on_error": "warn"
-      }
-    ],
-    "implementation_approach": [                   // REQUIRED: 3 TDD phases
-      {
-        "step": 1,
-        "title": "RED Phase: Write failing tests",
-        "tdd_phase": "red",                        // REQUIRED: Phase identifier
-        "description": "Write comprehensive failing tests",
-        "modification_points": ["Files/changes to make"],
-        "logic_flow": ["Step-by-step process"],
-        "acceptance": ["Phase success criteria"],
-        "depends_on": [],
-        "output": "failing_tests"
-      },
-      {
-        "step": 2,
-        "title": "GREEN Phase: Implement to pass tests",
-        "tdd_phase": "green",                      // REQUIRED: Phase identifier
-        "description": "Minimal implementation with test-fix cycle",
-        "modification_points": ["Implementation files"],
-        "logic_flow": [
-          "Implement minimal code",
-          "Run tests",
-          "If fail → Enter iteration loop (max 3):",
-          "  1. Extract failure messages",
-          "  2. Gemini bug-fix diagnosis",
-          "  3. Apply fixes",
-          "  4. Rerun tests",
-          "If max_iterations → Auto-revert"
-        ],
-        "acceptance": ["All tests pass"],
-        "command": "bash(npm test -- tests/path/)",
-        "depends_on": [1],
-        "output": "passing_implementation"
-      },
-      {
-        "step": 3,
-        "title": "REFACTOR Phase: Improve code quality",
-        "tdd_phase": "refactor",                   // REQUIRED: Phase identifier
-        "description": "Refactor while keeping tests green",
-        "modification_points": ["Quality improvements"],
-        "logic_flow": ["Incremental refactoring with test verification"],
-        "acceptance": ["Tests still pass", "Code quality improved"],
-        "command": "bash(npm run lint && npm test)",
-        "depends_on": [2],
-        "output": "refactored_implementation"
-      }
-    ],
-    "post_completion": [                           // OPTIONAL: Final verification
-      {
-        "step": "verify_full_tdd_cycle",
-        "action": "Confirm complete TDD cycle",
-        "command": "bash(npm test && echo 'TDD complete')",
-        "output_to": "final_validation",
-        "on_error": "fail"
-      }
-    ],
-    "error_handling": {                            // OPTIONAL: Error recovery
-      "green_phase_max_iterations": {
-        "action": "revert_all_changes",
-        "commands": ["bash(git reset --hard HEAD)"],
-        "report": "Generate failure report"
-      }
-    }
+  "context_package_path": ".workflow/active//{session-id}/.process/context-package.json",
+  "context_package": {
+    // If in memory: use cached content
+    // Else: Load from .workflow/active//{session-id}/.process/context-package.json
+  },
+  "test_context_package_path": ".workflow/active//{session-id}/.process/test-context-package.json",
+  "test_context_package": {
+    // Existing test patterns and coverage analysis
+  },
+  "mcp_capabilities": {
+    "codex_lens": true,
+    "exa_code": true,
+    "exa_web": true
   }
 }
 ```
 
-**Key JSON Fields Summary**:
-- `meta.tdd_workflow`: Must be `true`
-- `meta.max_iterations`: Green phase fix cycle limit (default: 3)
-- `meta.use_codex`: Automated fixes (false=manual, true=Codex)
-- `flow_control.implementation_approach`: Exactly 3 steps with `tdd_phase`: "red", "green", "refactor"
-- `context.tdd_cycles`: Optional detailed test cycle specifications
-- `context.parent`: Required for subtasks (IMPL-N.M)
+**Discovery Actions**:
+1. **Load Session Context** (if not in memory)
+   ```javascript
+   if (!memory.has("workflow-session.json")) {
+     Read(.workflow/active//{session-id}/workflow-session.json)
+   }
+   ```
 
-#### IMPL_PLAN.md Structure
+2. **Load Context Package** (if not in memory)
+   ```javascript
+   if (!memory.has("context-package.json")) {
+     Read(.workflow/active//{session-id}/.process/context-package.json)
+   }
+   ```
 
-Generate IMPL_PLAN.md with 8-section structure:
+3. **Load Test Context Package** (if not in memory)
+   ```javascript
+   if (!memory.has("test-context-package.json")) {
+     Read(.workflow/active//{session-id}/.process/test-context-package.json)
+   }
+   ```
 
-**Frontmatter** (required fields):
-```yaml
----
-identifier: WFS-{session-id}
-source: "User requirements" | "File: path"
-conflict_resolution: .workflow/{session-id}/.process/CONFLICT_RESOLUTION.md  # if exists
-context_package: .workflow/{session-id}/.process/context-package.json
-context_package_path: .workflow/{session-id}/.process/context-package.json
-test_context: .workflow/{session-id}/.process/test-context-package.json  # if exists
-workflow_type: "tdd"
-verification_history:
-  conflict_resolution: "executed | skipped" # based on conflict_risk
-  action_plan_verify: "pending"
-phase_progression: "brainstorm → context → test_context → conflict_resolution → tdd_planning"
-feature_count: N
-task_count: N  # ≤10 total
-task_breakdown:
-  simple_features: K
-  complex_features: L
-  total_subtasks: M
-tdd_workflow: true
----
+4. **Extract & Load Role Analyses** (from context-package.json)
+   ```javascript
+   // Extract role analysis paths from context package
+   const roleAnalysisPaths = contextPackage.brainstorm_artifacts.role_analyses
+     .flatMap(role => role.files.map(f => f.path));
+
+   // Load each role analysis file
+   roleAnalysisPaths.forEach(path => Read(path));
+   ```
+
+5. **Load Conflict Resolution** (from conflict-resolution.json, if exists)
+   ```javascript
+   // Check for new conflict-resolution.json format
+   if (contextPackage.conflict_detection?.resolution_file) {
+     Read(contextPackage.conflict_detection.resolution_file)  // .process/conflict-resolution.json
+   }
+   // Fallback: legacy brainstorm_artifacts path
+   else if (contextPackage.brainstorm_artifacts?.conflict_resolution?.exists) {
+     Read(contextPackage.brainstorm_artifacts.conflict_resolution.path)
+   }
+   ```
+
+6. **Code Analysis with Native Tools** (optional - enhance understanding)
+   ```bash
+   # Find relevant test files and patterns
+   find . -name "*test*" -type f
+   rg "describe|it\(|test\(" -g "*.ts"
+   ```
+
+7. **MCP External Research** (optional - gather TDD best practices)
+   ```javascript
+   // Get external TDD examples and patterns
+   mcp__exa__get_code_context_exa(
+     query="TypeScript TDD best practices Red-Green-Refactor",
+     tokensNum="dynamic"
+   )
+   ```
+
+### Phase 2: Agent Execution (Document Generation)
+
+**Pre-Agent Template Selection** (Command decides path before invoking agent):
+```javascript
+// Command checks flag and selects template PATH (not content)
+const templatePath = hasCliExecuteFlag
+  ? "~/.claude/workflows/cli-templates/prompts/workflow/task-json-cli-mode.txt"
+  : "~/.claude/workflows/cli-templates/prompts/workflow/task-json-agent-mode.txt";
 ```
 
-**8 Sections Structure**:
+**Agent Invocation**:
+```javascript
+Task(
+  subagent_type="action-planning-agent",
+  run_in_background=false,
+  description="Generate TDD task JSON and implementation plan",
+  prompt=`
+## Execution Context
 
-```markdown
-# Implementation Plan: {Project Title}
+**Session ID**: WFS-{session-id}
+**Workflow Type**: TDD
+**Note**: CLI tool usage is determined semantically from user's task description
 
-## 1. Summary
-- Core requirements and objectives (2-3 paragraphs)
-- TDD-specific technical approach
+## Phase 1: Discovery Results (Provided Context)
 
-## 2. Context Analysis
-- CCW Workflow Context (Phase progression, Quality gates)
-- Context Package Summary (Focus paths, Test context)
-- Project Profile (Type, Scale, Tech Stack, Timeline)
-- Module Structure (Directory tree)
-- Dependencies (Primary, Testing, Development)
-- Patterns & Conventions
+### Session Metadata
+{session_metadata_content}
 
-## 3. Brainstorming Artifacts Reference
-- Artifact Usage Strategy
-  - CONFLICT_RESOLUTION.md (if exists - selected resolution strategies)
-  - role analysis documents (primary reference)
-  - test-context-package.json (test patterns)
-  - context-package.json (smart context)
-- Artifact Priority in Development
+### Role Analyses (Enhanced by Synthesis)
+{role_analyses_content}
+- Includes requirements, design specs, enhancements, and clarifications from synthesis phase
 
-## 4. Implementation Strategy
-- Execution Strategy (TDD Cycles: Red-Green-Refactor)
-- Architectural Approach
-- Key Dependencies (Task dependency graph)
-- Testing Strategy (Coverage targets, Quality gates)
+### Artifacts Inventory
+- **Guidance Specification**: {guidance_spec_path}
+- **Role Analyses**: {role_analyses_list}
 
-## 5. TDD Implementation Tasks
-- Feature-by-Feature TDD Tasks
-  - Each task: IMPL-N with internal Red → Green → Refactor
-  - Dependencies and complexity metrics
-- Complex Feature Examples (when subtasks needed)
-- TDD Task Breakdown Summary
+### Context Package
+{context_package_summary}
+- Includes conflict_risk assessment
 
-## 6. Implementation Plan (Detailed Phased Breakdown)
-- Execution Strategy (feature-by-feature sequential)
-- Phase breakdown (Phase 1, Phase 2, etc.)
-- Resource Requirements (Team, Dependencies, Infrastructure)
+### Test Context Package
+{test_context_package_summary}
+- Existing test patterns, framework config, coverage analysis
 
-## 7. Risk Assessment & Mitigation
-- Risk table (Risk, Impact, Probability, Mitigation, Owner)
-- Critical Risks (TDD-specific)
-- Monitoring Strategy
+### Conflict Resolution (Conditional)
+If conflict_risk was medium/high, modifications have been applied to:
+- **guidance-specification.md**: Design decisions updated to resolve conflicts
+- **Role analyses (*.md)**: Recommendations adjusted for compatibility
+- **context-package.json**: Marked as "resolved" with conflict IDs
+- Conflict resolution results stored in conflict-resolution.json
 
-## 8. Success Criteria
-- Functional Completeness
-- Technical Quality (Test coverage ≥80%)
-- Operational Readiness
-- TDD Compliance
+### MCP Analysis Results (Optional)
+**Code Structure**: {mcp_code_index_results}
+**External Research**: {mcp_exa_research_results}
+
+## Phase 2: TDD Document Generation Task
+
+**Agent Configuration Reference**: All TDD task generation rules, quantification requirements, Red-Green-Refactor cycle structure, quality standards, and execution details are defined in action-planning-agent.
+
+### TDD-Specific Requirements Summary
+
+#### Task Structure Philosophy
+- **1 feature = 1 task** containing complete TDD cycle internally
+- Each task executes Red-Green-Refactor phases sequentially
+- Task count = Feature count (typically 5 features = 5 tasks)
+- Subtasks only when complexity >2500 lines or >6 files per cycle
+- **Maximum 18 tasks** (hard limit for TDD workflows)
+
+#### TDD Cycle Mapping
+- **Simple features**: IMPL-N with internal Red-Green-Refactor phases
+- **Complex features**: IMPL-N (container) + IMPL-N.M (subtasks)
+- Each cycle includes: test_count, test_cases array, implementation_scope, expected_coverage
+
+#### Required Outputs Summary
+
+##### 1. TDD Task JSON Files (.task/IMPL-*.json)
+- **Location**: `.workflow/active//{session-id}/.task/`
+- **Schema**: 5-field structure with TDD-specific metadata
+  - `meta.tdd_workflow`: true (REQUIRED)
+  - `meta.max_iterations`: 3 (Green phase test-fix cycle limit)
+  - `context.tdd_cycles`: Array with quantified test cases and coverage
+  - `flow_control.implementation_approach`: Exactly 3 steps with `tdd_phase` field
+    1. Red Phase (`tdd_phase: "red"`): Write failing tests
+    2. Green Phase (`tdd_phase: "green"`): Implement to pass tests
+    3. Refactor Phase (`tdd_phase: "refactor"`): Improve code quality
+  - CLI tool usage determined semantically (add `command` field when user requests CLI execution)
+- **Details**: See action-planning-agent.md § TDD Task JSON Generation
+
+##### 2. IMPL_PLAN.md (TDD Variant)
+- **Location**: `.workflow/active//{session-id}/IMPL_PLAN.md`
+- **Template**: `~/.claude/workflows/cli-templates/prompts/workflow/impl-plan-template.txt`
+- **TDD-Specific Frontmatter**: workflow_type="tdd", tdd_workflow=true, feature_count, task_breakdown
+- **TDD Implementation Tasks Section**: Feature-by-feature with internal Red-Green-Refactor cycles
+- **Details**: See action-planning-agent.md § TDD Implementation Plan Creation
+
+##### 3. TODO_LIST.md
+- **Location**: `.workflow/active//{session-id}/TODO_LIST.md`
+- **Format**: Hierarchical task list with internal TDD phase indicators (Red → Green → Refactor)
+- **Status**: ▸ (container), [ ] (pending), [x] (completed)
+- **Details**: See action-planning-agent.md § TODO List Generation
+
+### Quantification Requirements (MANDATORY)
+
+**Core Rules**:
+1. **Explicit Test Case Counts**: Red phase specifies exact number with enumerated list
+2. **Quantified Coverage**: Acceptance includes measurable percentage (e.g., ">=85%")
+3. **Detailed Implementation Scope**: Green phase enumerates files, functions, line counts
+4. **Enumerated Refactoring Targets**: Refactor phase lists specific improvements with counts
+
+**TDD Phase Formats**:
+- **Red Phase**: "Write N test cases: [test1, test2, ...]"
+- **Green Phase**: "Implement N functions in file lines X-Y: [func1() X1-Y1, func2() X2-Y2, ...]"
+- **Refactor Phase**: "Apply N refactorings: [improvement1 (details), improvement2 (details), ...]"
+- **Acceptance**: "All N tests pass with >=X% coverage: verify by [test command]"
+
+**Validation Checklist**:
+- [ ] Every Red phase specifies exact test case count with enumerated list
+- [ ] Every Green phase enumerates files, functions, and estimated line counts
+- [ ] Every Refactor phase lists specific improvements with counts
+- [ ] Every acceptance criterion includes measurable coverage percentage
+- [ ] tdd_cycles array contains test_count and test_cases for each cycle
+- [ ] No vague language ("comprehensive", "complete", "thorough")
+
+### Agent Execution Summary
+
+**Key Steps** (Detailed instructions in action-planning-agent.md):
+1. Load task JSON template from provided path
+2. Extract and decompose features with TDD cycles
+3. Generate TDD task JSON files enforcing quantification requirements
+4. Create IMPL_PLAN.md using TDD template variant
+5. Generate TODO_LIST.md with TDD phase indicators
+6. Update session state with TDD metadata
+
+**Quality Gates** (Full checklist in action-planning-agent.md):
+- ✓ Quantification requirements enforced (explicit counts, measurable acceptance, exact targets)
+- ✓ Task count ≤18 (hard limit)
+- ✓ Each task has meta.tdd_workflow: true
+- ✓ Each task has exactly 3 implementation steps with tdd_phase field
+- ✓ Green phase includes test-fix cycle logic
+- ✓ Artifact references mapped correctly
+- ✓ MCP tool integration added
+- ✓ Documents follow TDD template structure
+
+## Output
+
+Generate all three documents and report completion status:
+- TDD task JSON files created: N files (IMPL-*.json)
+- TDD cycles configured: N cycles with quantified test cases
+- Artifacts integrated: synthesis-spec, guidance-specification, N role analyses
+- Test context integrated: existing patterns and coverage
+- MCP enhancements: CodexLens, exa-research
+- Session ready for TDD execution: /workflow:execute
+`
+)
 ```
 
-### Phase 4: TODO_LIST.md Generation
+### Agent Context Passing
 
-Generate task list with internal TDD phase indicators:
+**Memory-Aware Context Assembly**:
+```javascript
+// Assemble context package for agent
+const agentContext = {
+  session_id: "WFS-[id]",
+  workflow_type: "tdd",
 
-**For Simple Features (1 task per feature)**:
-```markdown
-## TDD Implementation Tasks
+  // Use memory if available, else load
+  session_metadata: memory.has("workflow-session.json")
+    ? memory.get("workflow-session.json")
+    : Read(.workflow/active/WFS-[id]/workflow-session.json),
 
-### Feature 1: {Feature Name}
-- [ ] **IMPL-1**: Implement {feature} with TDD → [Task](./.task/IMPL-1.json)
-  - Internal phases: Red → Green → Refactor
-  - Dependencies: None
+  context_package_path: ".workflow/active/WFS-[id]/.process/context-package.json",
 
-### Feature 2: {Feature Name}
-- [ ] **IMPL-2**: Implement {feature} with TDD → [Task](./.task/IMPL-2.json)
-  - Internal phases: Red → Green → Refactor
-  - Dependencies: IMPL-1
-```
+  context_package: memory.has("context-package.json")
+    ? memory.get("context-package.json")
+    : Read(".workflow/active/WFS-[id]/.process/context-package.json"),
 
-**For Complex Features (with subtasks)**:
-```markdown
-### Feature 3: {Complex Feature Name}
-▸ **IMPL-3**: Implement {complex feature} with TDD → [Task](./.task/IMPL-3.json)
-  - [ ] **IMPL-3.1**: {Sub-feature A} with TDD → [Task](./.task/IMPL-3.1.json)
-    - Internal phases: Red → Green → Refactor
-  - [ ] **IMPL-3.2**: {Sub-feature B} with TDD → [Task](./.task/IMPL-3.2.json)
-    - Internal phases: Red → Green → Refactor
-    - Dependencies: IMPL-3.1
-```
+  test_context_package_path: ".workflow/active/WFS-[id]/.process/test-context-package.json",
 
-**Status Legend**:
-```markdown
-## Status Legend
-- ▸ = Container task (has subtasks)
-- [ ] = Pending task
-- [x] = Completed task
-- Red = Write failing tests
-- Green = Implement to pass tests (with test-fix cycle)
-- Refactor = Improve code quality
-```
+  test_context_package: memory.has("test-context-package.json")
+    ? memory.get("test-context-package.json")
+    : Read(".workflow/active/WFS-[id]/.process/test-context-package.json"),
 
-### Phase 5: Session State Update
+  // Extract brainstorm artifacts from context package
+  brainstorm_artifacts: extractBrainstormArtifacts(context_package),
 
-Update workflow-session.json with TDD metadata:
-```json
-{
-  "workflow_type": "tdd",
-  "feature_count": 5,
-  "task_count": 5,
-  "task_breakdown": {
-    "simple_features": 4,
-    "complex_features": 1,
-    "total_subtasks": 2
-  },
-  "tdd_workflow": true,
-  "task_limit_compliance": true
+  // Load role analyses using paths from context package
+  role_analyses: brainstorm_artifacts.role_analyses
+    .flatMap(role => role.files)
+    .map(file => Read(file.path)),
+
+  // Load conflict resolution if exists (prefer new JSON format)
+  conflict_resolution: context_package.conflict_detection?.resolution_file
+    ? Read(context_package.conflict_detection.resolution_file)  // .process/conflict-resolution.json
+    : (brainstorm_artifacts?.conflict_resolution?.exists
+        ? Read(brainstorm_artifacts.conflict_resolution.path)
+        : null),
+
+  // Optional MCP enhancements
+  mcp_analysis: executeMcpDiscovery()
 }
 ```
 
-**Task Count Calculation**:
-- **Simple features**: 1 task each (IMPL-N with internal TDD cycle)
-- **Complex features**: 1 container + M subtasks (IMPL-N + IMPL-N.M)
-- **Total**: Simple feature count + Complex feature subtask count
-- **Example**: 4 simple + 1 complex (with 2 subtasks) = 6 total tasks (not 15)
+## TDD Task Structure Reference
+
+This section provides quick reference for TDD task JSON structure. For complete implementation details, see the agent invocation prompt in Phase 2 above.
+
+**Quick Reference**:
+- Each TDD task contains complete Red-Green-Refactor cycle
+- Task ID format: `IMPL-N` (simple) or `IMPL-N.M` (complex subtasks)
+- Required metadata: `meta.tdd_workflow: true`, `meta.max_iterations: 3`
+- Flow control: Exactly 3 steps with `tdd_phase` field (red, green, refactor)
+- Context: `tdd_cycles` array with quantified test cases and coverage
+- See Phase 2 agent prompt for full schema and requirements
 
 ## Output Files Structure
 ```
-.workflow/{session-id}/
+.workflow/active//{session-id}/
 ├── IMPL_PLAN.md                     # Unified plan with TDD Implementation Tasks section
 ├── TODO_LIST.md                     # Progress tracking with internal TDD phase indicators
 ├── .task/
@@ -411,7 +407,7 @@ Update workflow-session.json with TDD metadata:
 │   ├── IMPL-3.2.json                # Complex feature subtask (if needed)
 │   └── ...
 └── .process/
-    ├── CONFLICT_RESOLUTION.md       # Conflict resolution strategies (if conflict_risk ≥ medium)
+    ├── conflict-resolution.json     # Conflict resolution results (if conflict_risk ≥ medium)
     ├── test-context-package.json    # Test coverage analysis
     ├── context-package.json         # Input from context-gather
     ├── context_package_path         # Path to smart context package
@@ -465,52 +461,28 @@ Update workflow-session.json with TDD metadata:
 
 ## Integration & Usage
 
-### Command Chain
-- **Called By**: `/workflow:tdd-plan` (Phase 4)
-- **Calls**: Gemini CLI for TDD breakdown
-- **Followed By**: `/workflow:execute`, `/workflow:tdd-verify`
+**Command Chain**:
+- Called by: `/workflow:tdd-plan` (Phase 4)
+- Invokes: `action-planning-agent` for autonomous task generation
+- Followed by: `/workflow:execute`, `/workflow:tdd-verify`
 
-### Basic Usage
+**Basic Usage**:
 ```bash
-# Manual mode (default)
+# Standard execution
 /workflow:tools:task-generate-tdd --session WFS-auth
 
-# Agent mode (autonomous task generation)
-/workflow:tools:task-generate-tdd --session WFS-auth --agent
+# With semantic CLI request (include in task description)
+# e.g., "Generate TDD tasks for auth module, use Codex for implementation"
 ```
 
-### Expected Output
-```
-TDD task generation complete for session: WFS-auth
+**CLI Tool Selection**: Determined semantically from user's task description. Include "use Codex/Gemini/Qwen" in your request for CLI execution.
 
-Features analyzed: 5
-Total tasks: 5 (1 task per feature with internal TDD cycles)
-
-Task breakdown:
-- Simple features: 4 tasks (IMPL-1 to IMPL-4)
-- Complex features: 1 task with 2 subtasks (IMPL-5, IMPL-5.1, IMPL-5.2)
-- Total task count: 6 (within 10-task limit)
-
-Structure:
-- IMPL-1: User Authentication (Internal: Red → Green → Refactor)
-- IMPL-2: Password Reset (Internal: Red → Green → Refactor)
-- IMPL-3: Email Verification (Internal: Red → Green → Refactor)
-- IMPL-4: Role Management (Internal: Red → Green → Refactor)
-- IMPL-5: Payment System (Container)
-  - IMPL-5.1: Gateway Integration (Internal: Red → Green → Refactor)
-  - IMPL-5.2: Transaction Management (Internal: Red → Green → Refactor)
-
-Plans generated:
-- Unified Plan: .workflow/WFS-auth/IMPL_PLAN.md (includes TDD Implementation Tasks section)
-- Task List: .workflow/WFS-auth/TODO_LIST.md (with internal TDD phase indicators)
-
-TDD Configuration:
-- Each task contains complete Red-Green-Refactor cycle
-- Green phase includes test-fix cycle (max 3 iterations)
-- Auto-revert on max iterations reached
-
-Next: /workflow:action-plan-verify --session WFS-auth (recommended) or /workflow:execute --session WFS-auth
-```
+**Output**:
+- TDD task JSON files in `.task/` directory (IMPL-N.json format)
+- IMPL_PLAN.md with TDD Implementation Tasks section
+- TODO_LIST.md with internal TDD phase indicators
+- Session state updated with task count and TDD metadata
+- MCP enhancements integrated (if available)
 
 ## Test Coverage Analysis Integration
 
@@ -532,24 +504,14 @@ IMPL (Green phase) tasks include automatic test-fix cycle:
 3. **Success Path**: Tests pass → Complete task
 4. **Failure Path**: Tests fail → Enter iterative fix cycle:
    - **Gemini Diagnosis**: Analyze failures with bug-fix template
-   - **Fix Application**: Manual (default) or Codex (if meta.use_codex=true)
+   - **Fix Application**: Agent (default) or CLI (if `command` field present)
    - **Retest**: Verify fix resolves failures
    - **Repeat**: Up to max_iterations (default: 3)
 5. **Safety Net**: Auto-revert all changes if max iterations reached
 
-**Key Benefits**:
-- Faster feedback loop within Green phase
-- Autonomous recovery from initial implementation errors
-- Systematic debugging with Gemini's bug-fix template
-- Safe rollback prevents broken TDD state
+
 
 ## Configuration Options
 - **meta.max_iterations**: Number of fix attempts (default: 3 for TDD, 5 for test-gen)
-- **meta.use_codex**: Enable Codex automated fixes (default: false, manual)
+- **CLI tool usage**: Determined semantically from user's task description via `command` field in implementation_approach
 
-## Related Commands
-- `/workflow:tdd-plan` - Orchestrates TDD workflow planning (6 phases)
-- `/workflow:tools:test-context-gather` - Analyzes test coverage
-- `/workflow:execute` - Executes TDD tasks in order
-- `/workflow:tdd-verify` - Verifies TDD compliance
-- `/workflow:test-gen` - Post-implementation test generation

.claude/commands/workflow/tools/task-generate.md

@@ -1,815 +0,0 @@
---- 
-name: task-generate
-description: Generate task JSON files and IMPL_PLAN.md from analysis results with artifacts integration
-argument-hint: "--session WFS-session-id [--cli-execute]"
-examples:
-  - /workflow:tools:task-generate --session WFS-auth
-  - /workflow:tools:task-generate --session WFS-auth --cli-execute
----
-
-# Task Generation Command
-
-## 1. Overview
-This command generates task JSON files and an `IMPL_PLAN.md` from brainstorming role analyses. It automatically detects and integrates all brainstorming artifacts (role-specific `analysis.md` files and `guidance-specification.md`), creating a structured and context-rich plan for implementation. The command supports two primary execution modes: a default agent-based mode for seamless context handling and a `--cli-execute` mode that leverages the Codex CLI for complex, autonomous development tasks. Its core function is to translate requirements and design specifications from role analyses into actionable, executable tasks, ensuring all necessary context, dependencies, and implementation steps are defined upfront.
-
-## 2. Execution Modes
-
-This command offers two distinct modes for task execution, providing flexibility for different implementation complexities.
-
-### Agent Mode (Default)
-In the default mode, each step in `implementation_approach` **omits the `command` field**. The agent interprets the step's `modification_points` and `logic_flow` to execute the task autonomously.
-- **Step Structure**: Contains `step`, `title`, `description`, `modification_points`, `logic_flow`, `depends_on`, and `output` fields
-- **Execution**: Agent reads these fields and performs the implementation autonomously
-- **Context Loading**: Agent loads context via `pre_analysis` steps
-- **Validation**: Agent validates against acceptance criteria in `context.acceptance`
-- **Benefit**: Direct agent execution with full context awareness, no external tool overhead
-- **Use Case**: Standard implementation tasks where agent capability is sufficient
-
-### CLI Execute Mode (`--cli-execute`)
-When the `--cli-execute` flag is used, each step in `implementation_approach` **includes a `command` field** that specifies the exact execution command. This mode is designed for complex implementations requiring specialized CLI tools.
-- **Step Structure**: Includes all default fields PLUS a `command` field
-- **Execution**: The specified command executes the step directly (e.g., `bash(codex ...)`)
-- **Context Packages**: Each command receives context via the CONTEXT field in the prompt
-- **Multi-Step Support**: Complex tasks can have multiple sequential codex steps with `resume --last`
-- **Benefit**: Leverages specialized CLI tools (codex/gemini/qwen) for complex reasoning and autonomous execution
-- **Use Case**: Large-scale features, complex refactoring, or when user explicitly requests CLI tool usage
-
-## 3. Core Principles
-This command is built on a set of core principles to ensure efficient and reliable task generation.
-
-- **Role Analysis-Driven**: All generated tasks originate from role-specific `analysis.md` files (enhanced in synthesis phase), ensuring direct link between requirements/design and implementation
-- **Artifact-Aware**: Automatically detects and integrates all brainstorming outputs (role analyses, guidance-specification.md, enhancements) to enrich task context
-- **Context-Rich**: Embeds comprehensive context (requirements, focus paths, acceptance criteria, artifact references) directly into each task JSON
-- **Path Clarity**: All `focus_paths` prefer absolute paths (e.g., `D:\\project\\src\\module`), or clear relative paths from project root (e.g., `./src/module`)
-- **Flow-Control Ready**: Pre-defines clear execution sequence (`pre_analysis`, `implementation_approach`) within each task
-- **Memory-First**: Prioritizes using documents already loaded in conversation memory to avoid redundant file operations
-- **Mode-Flexible**: Supports both agent-driven execution (default) and CLI tool execution (with `--cli-execute` flag)
-- **Multi-Step Support**: Complex tasks can use multiple sequential steps in `implementation_approach` with codex resume mechanism
-- **Responsibility**: Parses analysis, detects artifacts, generates enhanced task JSONs, creates `IMPL_PLAN.md` and `TODO_LIST.md`, updates session state
-
-## 4. Execution Flow
-The command follows a streamlined, three-step process to convert analysis into executable tasks.
-
-### Step 1: Input & Discovery
-The process begins by gathering all necessary inputs. It follows a **Memory-First Rule**, skipping file reads if documents are already in the conversation memory.
-1.  **Session Validation**: Loads and validates the session from `.workflow/{session_id}/workflow-session.json`.
-2.  **Context Package Loading** (primary source): Reads `.workflow/{session_id}/.process/context-package.json` for smart context and artifact catalog.
-3.  **Brainstorm Artifacts Extraction**: Extracts role analysis paths from `context-package.json` → `brainstorm_artifacts.role_analyses[]` (supports `analysis*.md` automatically).
-4.  **Document Loading**: Reads role analyses, guidance specification, synthesis output, and conflict resolution (if exists) using paths from context package.
-
-### Step 2: Task Decomposition & Grouping
-Once all inputs are loaded, the command analyzes the tasks defined in the analysis results and groups them based on shared context.
-1.  **Task Definition Parsing**: Extracts task definitions, requirements, and dependencies.
-2.  **Context Signature Analysis**: Computes a unique hash (`context_signature`) for each task based on its `focus_paths` and referenced `artifacts`.
-3.  **Task Grouping**:
-    *   Tasks with the **same signature** are candidates for merging, as they operate on the same context.
-    *   Tasks with **different signatures** and no dependencies are grouped for parallel execution.
-    *   Tasks with `depends_on` relationships are marked for sequential execution.
-4.  **Modification Target Determination**: Extracts specific code locations (`file:function:lines`) from the analysis to populate the `target_files` field.
-
-### Step 3: Output Generation
-Finally, the command generates all the necessary output files.
-1.  **Task JSON Creation**: Creates individual `.task/IMPL-*.json` files, embedding all context, artifacts, and flow control steps. If `--cli-execute` is active, it generates the appropriate `codex exec` commands.
-2.  **IMPL_PLAN.md Generation**: Creates the main implementation plan document, summarizing the strategy, tasks, and dependencies.
-3.  **TODO_LIST.md Generation**: Creates a simple checklist for tracking task progress.
-4.  **Session State Update**: Updates `workflow-session.json` with the final task count and artifact inventory, marking the session as ready for execution.
-
-## 5. Task Decomposition Strategy
-The command employs a sophisticated strategy to group and decompose tasks, optimizing for context reuse and parallel execution.
-
-### Core Principles
-- **Primary Rule: Shared Context → Merge Tasks**: Tasks that operate on the same files, use the same artifacts, and share the same tech stack are merged. This avoids redundant context loading and recognizes inherent relationships between the tasks.
-- **Secondary Rule: Different Contexts + No Dependencies → Decompose for Parallel Execution**: Tasks that are fully independent (different files, different artifacts, no shared dependencies) are decomposed into separate parallel execution groups.
-
-### Context Analysis for Task Grouping
-The decision to merge or decompose is based on analyzing context indicators:
-
-1.  **Shared Context Indicators (→ Merge)**:
-    *   Identical `focus_paths` (working on the same modules/files).
-    *   Same tech stack and dependencies.
-    *   Identical `context.artifacts` references.
-    *   A sequential logic flow within the same feature.
-    *   Shared test fixtures or setup.
-
-2.  **Independent Context Indicators (→ Decompose)**:
-    *   Different `focus_paths` (separate modules).
-    *   Different tech stacks (e.g., frontend vs. backend).
-    *   Different `context.artifacts` (using different brainstorming outputs).
-    *   No shared dependencies.
-    *   Can be tested independently.
-
-**Decomposition is only performed when**:
-- Tasks have different contexts and no shared dependencies (enabling parallel execution).
-- A single task represents an excessive workload (e.g., >2500 lines of code or >6 files to modify).
-- A sequential dependency creates a necessary block (e.g., IMPL-1 must complete before IMPL-2 can start).
-
-### Context Signature Algorithm
-To automate grouping, a `context_signature` is computed for each task.
-
-```javascript
-// Compute context signature for task grouping
-function computeContextSignature(task) {
-  const focusPathsStr = task.context.focus_paths.sort().join('|');
-  const artifactsStr = task.context.artifacts.map(a => a.path).sort().join('|');
-  const techStack = task.context.shared_context?.tech_stack?.sort().join('|') || '';
-
-  return hash(`${focusPathsStr}:${artifactsStr}:${techStack}`);
-}
-```
-
-### Execution Group Assignment
-Tasks are assigned to execution groups based on their signatures and dependencies.
-
-```javascript
-// Group tasks by context signature
-function groupTasksByContext(tasks) {
-  const groups = {};
-
-  tasks.forEach(task => {
-    const signature = computeContextSignature(task);
-    if (!groups[signature]) {
-      groups[signature] = [];
-    }
-    groups[signature].push(task);
-  });
-
-  return groups;
-}
-
-// Assign execution groups for parallel tasks
-function assignExecutionGroups(tasks) {
-  const contextGroups = groupTasksByContext(tasks);
-
-  Object.entries(contextGroups).forEach(([signature, groupTasks]) => {
-    if (groupTasks.length === 1) {
-      const task = groupTasks[0];
-      // Single task with unique context
-      if (!task.context.depends_on || task.context.depends_on.length === 0) {
-        task.meta.execution_group = `parallel-${signature.slice(0, 8)}`;
-      } else {
-        task.meta.execution_group = null; // Sequential task
-      }
-    } else {
-      // Multiple tasks with same context → Should be merged
-      console.warn(`Tasks ${groupTasks.map(t => t.id).join(', ')} share context and should be merged`);
-      // Merge tasks into single task
-      return mergeTasks(groupTasks);
-    }
-  });
-}
-```
-**Task Limits**:
-- **Maximum 10 tasks** (hard limit).
-- **Hierarchy**: Flat (≤5 tasks) or two-level (6-10 tasks). If >10, the scope should be re-evaluated.
-- **Parallel Groups**: Tasks with the same `execution_group` ID are independent and can run concurrently.
-
-## 6. Generated Outputs
-The command produces three key documents and a directory of task files.
-
-### 6.1. Task JSON Schema (`.task/IMPL-*.json`)
-This enhanced 5-field schema embeds all necessary context, artifacts, and execution steps.
-
-```json
-{
-  "id": "IMPL-N[.M]",
-  "title": "Descriptive task name",
-  "status": "pending|active|completed|blocked|container",
-  "context_package_path": ".workflow/WFS-[session]/.process/context-package.json",
-  "meta": {
-    "type": "feature|bugfix|refactor|test-gen|test-fix|docs",
-    "agent": "@code-developer|@test-fix-agent|@universal-executor",
-    "execution_group": "group-id|null",
-    "context_signature": "hash-of-focus_paths-and-artifacts"
-  },
-  "context": {
-    "requirements": ["Clear requirement from analysis"],
-    "focus_paths": ["D:\\project\\src\\module\\path", "./tests/module/path"],
-    "acceptance": ["Measurable acceptance criterion"],
-    "parent": "IMPL-N",
-    "depends_on": ["IMPL-N.M"],
-    "inherited": {"shared_patterns": [], "common_dependencies": []},
-    "shared_context": {"tech_stack": [], "conventions": []},
-    "artifacts": [
-      {
-        "path": "{{from context-package.json → brainstorm_artifacts.role_analyses[].files[].path}}",
-        "priority": "highest",
-        "usage": "Role-specific requirements, design specs, enhanced by synthesis. Paths loaded dynamically from context-package.json (supports multiple files per role: analysis.md, analysis-01.md, analysis-api.md, etc.). Common roles: product-manager, system-architect, ui-designer, data-architect, ux-expert."
-      },
-      {
-        "path": ".workflow/WFS-[session]/.brainstorming/guidance-specification.md",
-        "priority": "high",
-        "usage": "Finalized design decisions (potentially modified by conflict resolution if conflict_risk was medium/high). Use for: understanding resolved requirements, design choices, conflict resolutions applied in-place"
-      }
-    ]
-  },
-  "flow_control": {
-    "pre_analysis": [
-      {
-        "step": "load_context_package",
-        "action": "Load context package for artifact paths",
-        "note": "Context package path is now at top-level field: context_package_path",
-        "commands": [
-          "Read({{context_package_path}})"
-        ],
-        "output_to": "context_package",
-        "on_error": "fail"
-      },
-      {
-        "step": "load_role_analysis_artifacts",
-        "action": "Load role analyses from context-package.json (supports multiple files per role)",
-        "note": "Paths loaded from context-package.json → brainstorm_artifacts.role_analyses[]. Supports analysis*.md automatically.",
-        "commands": [
-          "Read({{context_package_path}})",
-          "Extract(brainstorm_artifacts.role_analyses[].files[].path)",
-          "Read(each extracted path)"
-        ],
-        "output_to": "role_analysis_artifacts",
-        "on_error": "skip_optional"
-      },
-      {
-        "step": "load_planning_context",
-        "action": "Load plan-generated context intelligence with resolved conflicts",
-        "note": "CRITICAL: context-package.json (from context_package_path) provides smart context (focus paths, dependencies, patterns) and conflict resolution status. If conflict_risk was medium/high, conflicts have been resolved in guidance-specification.md and role analyses.",
-        "commands": [
-          "Read({{context_package_path}})",
-          "Read(.workflow/WFS-[session]/.brainstorming/guidance-specification.md)"
-        ],
-        "output_to": "planning_context",
-        "on_error": "fail",
-        "usage_guidance": {
-          "context-package.json": "Use for focus_paths validation, dependency resolution, existing pattern discovery, module structure understanding, conflict_risk status (resolved/none/low)",
-          "guidance-specification.md": "Use for finalized design decisions (includes applied conflict resolutions if any)"
-        }
-      },
-      {
-        "step": "codebase_exploration",
-        "action": "Explore codebase using native tools",
-        "command": "bash(find . -name \"[patterns]\" -type f && rg \"[patterns]\")",
-        "output_to": "codebase_structure"
-      },
-      {
-        "step": "analyze_task_patterns",
-        "action": "Analyze existing code patterns and identify modification targets",
-        "commands": [
-          "bash(cd \"[focus_paths]\")",
-          "bash(gemini \"PURPOSE: Identify modification targets TASK: Analyze '[title]' and locate specific files/functions/lines to modify CONTEXT: [role_analyses] [individual_artifacts] EXPECTED: Code locations in format 'file:function:lines' RULES: Consult role analyses for requirements, identify exact modification points\")"
-        ],
-        "output_to": "task_context_with_targets",
-        "on_error": "fail"
-      }
-    ],
-    "implementation_approach": [
-      {
-        "step": 1,
-        "title": "Implement task following role analyses and context",
-        "description": "Implement '[title]' following this priority: 1) role analysis.md files (requirements, design specs, enhancements from synthesis), 2) guidance-specification.md (finalized decisions with resolved conflicts), 3) context-package.json (smart context, focus paths, patterns). Role analyses are enhanced by synthesis phase with concept improvements and clarifications. If conflict_risk was medium/high, conflict resolutions are already applied in-place.",
-        "modification_points": [
-          "Apply requirements and design specs from role analysis documents",
-          "Use enhancements and clarifications from synthesis phase",
-          "Use finalized decisions from guidance-specification.md (includes resolved conflicts)",
-          "Use context-package.json for focus paths and dependency resolution",
-          "Consult specific role artifacts for implementation details when needed",
-          "Integrate with existing patterns"
-        ],
-        "logic_flow": [
-          "Load role analyses (requirements, design, enhancements from synthesis)",
-          "Load guidance-specification.md (finalized decisions with resolved conflicts if any)",
-          "Load context-package.json (smart context: focus paths, dependencies, patterns, conflict_risk status)",
-          "Extract requirements and design decisions from role documents",
-          "Review synthesis enhancements and clarifications",
-          "Use finalized decisions (conflicts already resolved if applicable)",
-          "Identify modification targets using context package",
-          "Implement following role requirements and design specs",
-          "Consult role artifacts for detailed specifications when needed",
-          "Validate against acceptance criteria"
-        ],
-        "depends_on": [],
-        "output": "implementation"
-      }
-    ],
-    "target_files": ["file:function:lines"]
-  }
-}
-```
-
-### 6.2. IMPL_PLAN.md Structure
-This document provides a high-level overview of the entire implementation plan.
-
-```markdown
----
-identifier: WFS-{session-id}
-source: "User requirements" | "File: path" | "Issue: ISS-001"
-role_analyses: .workflow/{session-id}/.brainstorming/[role]/analysis*.md
-artifacts: .workflow/{session-id}/.brainstorming/
-context_package: .workflow/{session-id}/.process/context-package.json  # CCW smart context
-guidance_specification: .workflow/{session-id}/.brainstorming/guidance-specification.md  # Finalized decisions with resolved conflicts
-workflow_type: "standard | tdd | design"  # Indicates execution model
-verification_history:  # CCW quality gates
-  synthesis_clarify: "passed | skipped | pending"  # Brainstorm phase clarification
-  action_plan_verify: "pending"
-  conflict_resolution: "resolved | none | low"  # Status from context-package.json
-phase_progression: "brainstorm → synthesis → context → conflict_resolution (if needed) → planning"  # CCW workflow phases
----
-
-# Implementation Plan: {Project Title}
-
-## 1. Summary
-Core requirements, objectives, technical approach summary (2-3 paragraphs max).
-
-**Core Objectives**:
-- [Key objective 1]
-- [Key objective 2]
-
-**Technical Approach**:
-- [High-level approach]
-
-## 2. Context Analysis
-
-### CCW Workflow Context
-**Phase Progression**:
-- ✅ Phase 1: Brainstorming (role analyses generated by participating roles)
-- ✅ Phase 2: Synthesis (concept enhancement + clarification, {N} questions answered, role analyses refined)
-- ✅ Phase 3: Context Gathering (context-package.json: {N} files, {M} modules analyzed, conflict_risk: {level})
-- ✅ Phase 4: Conflict Resolution ({status}: {conflict_count} conflicts detected and resolved | skipped if no conflicts)
-- ⏳ Phase 5: Task Generation (current phase - generating IMPL_PLAN.md and task JSONs)
-
-**Quality Gates**:
-- synthesis-clarify: ✅ Passed ({N} ambiguities resolved, {M} enhancements applied)
-- action-plan-verify: ⏳ Pending (recommended before /workflow:execute)
-
-**Context Package Summary**:
-- **Focus Paths**: {list key directories from context-package.json}
-- **Key Files**: {list primary files for modification}
-- **Module Depth Analysis**: {from get_modules_by_depth.sh output}
-- **Smart Context**: {total file count} files, {module count} modules, {dependency count} dependencies identified
-
-### Project Profile
-- **Type**: Greenfield/Enhancement/Refactor
-- **Scale**: User count, data volume, complexity
-- **Tech Stack**: Primary technologies
-- **Timeline**: Duration and milestones
-
-### Module Structure
-'''
-[Directory tree showing key modules]
-'''
-
-### Dependencies
-**Primary**: [Core libraries and frameworks]
-**APIs**: [External services]
-**Development**: [Testing, linting, CI/CD tools]
-
-### Patterns & Conventions
-- **Architecture**: [Key patterns like DI, Event-Driven]
-- **Component Design**: [Design patterns]
-- **State Management**: [State strategy]
-- **Code Style**: [Naming, TypeScript coverage]
-
-## 3. Brainstorming Artifacts Reference
-
-### Artifact Usage Strategy
-**Primary Reference (Role Analyses)**:
-- **What**: Role-specific analyses from brainstorming phase providing multi-perspective insights
-- **When**: Every task references relevant role analyses for requirements and design decisions
-- **How**: Extract requirements, architecture decisions, UI/UX patterns from applicable role documents
-- **Priority**: Collective authoritative source - multiple role perspectives provide comprehensive coverage
-- **CCW Value**: Maintains role-specific expertise while enabling cross-role integration during planning
-
-**Context Intelligence (context-package.json)**:
-- **What**: Smart context gathered by CCW's context-gather phase
-- **Content**: Focus paths, dependency graph, existing patterns, module structure, tech stack, conflict_risk status
-- **Usage**: Tasks load this via `flow_control.preparatory_steps` for environment setup and conflict awareness
-- **CCW Value**: Automated intelligent context discovery replacing manual file exploration
-
-**Conflict Resolution Status**:
-- **What**: Conflict resolution applied in-place to brainstorm artifacts (if conflict_risk was >= medium)
-- **Location**: guidance-specification.md and role analyses (*.md) contain resolved conflicts
-- **Status**: Check context-package.json → conflict_detection.conflict_risk ("resolved" | "none" | "low")
-- **Usage**: Read finalized decisions from guidance-specification.md (includes applied resolutions)
-- **CCW Value**: Interactive conflict resolution with user confirmation, modifications applied automatically
-
-### Role Analysis Documents (Highest Priority)
-Role analyses provide specialized perspectives on the implementation:
-- **system-architect/analysis.md**: Architecture design, ADRs, API specifications, caching strategies
-- **ui-designer/analysis.md**: Design tokens, layout specifications, component patterns
-- **ux-expert/analysis.md**: User journeys, interaction flows, accessibility requirements
-- **guidance-specification/analysis.md**: Product vision, user stories, business requirements, success metrics
-- **data-architect/analysis.md**: Data models, schemas, database design, migration strategies
-- **api-designer/analysis.md**: API contracts, endpoint specifications, integration patterns
-
-### Supporting Artifacts (Reference)
-- **topic-framework.md**: Role-specific discussion points and analysis framework
-
-**Artifact Priority in Development**:
-1. {context_package_path} (primary source: smart context AND brainstorm artifact catalog in `brainstorm_artifacts` + conflict_risk status)
-2. role/analysis*.md (paths from context-package.json: requirements, design specs, enhanced by synthesis, with resolved conflicts if any)
-3. guidance-specification.md (path from context-package.json: finalized decisions with resolved conflicts if any)
-
-## 4. Implementation Strategy
-
-### Execution Strategy
-**Execution Model**: [Sequential | Parallel | Phased | TDD Cycles]
-
-**Rationale**: [Why this execution model fits the project]
-
-**Parallelization Opportunities**:
-- [List independent workstreams]
-
-**Serialization Requirements**:
-- [List critical dependencies]
-
-### Architectural Approach
-**Key Architecture Decisions**:
-- [ADR references from role analyses]
-- [Justification for architecture patterns]
-
-**Integration Strategy**:
-- [How modules communicate]
-- [State management approach]
-
-### Key Dependencies
-**Task Dependency Graph**:
-'''
-[High-level dependency visualization]
-'''
-
-**Critical Path**: [Identify bottleneck tasks]
-
-### Testing Strategy
-**Testing Approach**:
-- Unit testing: [Tools, scope]
-- Integration testing: [Key integration points]
-- E2E testing: [Critical user flows]
-
-**Coverage Targets**:
-- Lines: ≥70%
-- Functions: ≥70%
-- Branches: ≥65%
-
-**Quality Gates**:
-- [CI/CD gates]
-- [Performance budgets]
-
-## 5. Task Breakdown Summary
-
-### Task Count
-**{N} tasks** (flat hierarchy | two-level hierarchy, sequential | parallel execution)
-
-### Task Structure
-- **IMPL-1**: [Main task title]
-- **IMPL-2**: [Main task title]
-...
-
-### Complexity Assessment
-- **High**: [List with rationale]
-- **Medium**: [List]
-- **Low**: [List]
-
-### Dependencies
-[Reference Section 4.3 for dependency graph]
-
-**Parallelization Opportunities**:
-- [Specific task groups that can run in parallel]
-
-## 6. Implementation Plan (Detailed Phased Breakdown)
-
-### Execution Strategy
-
-**Phase 1 (Weeks 1-2): [Phase Name]**
-- **Tasks**: IMPL-1, IMPL-2
-- **Deliverables**:
-  - [Specific deliverable 1]
-  - [Specific deliverable 2]
-- **Success Criteria**:
-  - [Measurable criterion]
-
-**Phase 2 (Weeks 3-N): [Phase Name]**
-...
-
-### Resource Requirements
-
-**Development Team**:
-- [Team composition and skills]
-
-**External Dependencies**:
-- [Third-party services, APIs]
-
-**Infrastructure**:
-- [Development, staging, production environments]
-
-## 7. Risk Assessment & Mitigation
-
-| Risk | Impact | Probability | Mitigation Strategy | Owner |
-|------|--------|-------------|---------------------|-------|
-| [Risk description] | High/Med/Low | High/Med/Low | [Strategy] | [Role] |
-
-**Critical Risks** (High impact + High probability):
-- [Risk 1]: [Detailed mitigation plan]
-
-**Monitoring Strategy**:
-- [How risks will be monitored]
-
-## 8. Success Criteria
-
-**Functional Completeness**:
-- [ ] All requirements from role analysis documents implemented
-- [ ] All acceptance criteria from task.json files met
-
-**Technical Quality**:
-- [ ] Test coverage ≥70%
-- [ ] Bundle size within budget
-- [ ] Performance targets met
-
-**Operational Readiness**:
-- [ ] CI/CD pipeline operational
-- [ ] Monitoring and logging configured
-- [ ] Documentation complete
-
-**Business Metrics**:
-- [ ] [Key business metrics from role analyses]
-```
-
-### 6.3. TODO_LIST.md Structure
-A simple Markdown file for tracking the status of each task.
-
-```markdown
-# Tasks: [Session Topic]
-
-## Task Progress
-▸ **IMPL-001**: [Main Task Group] → [📋](./.task/IMPL-001.json)
-  - [ ] **IMPL-001.1**: [Subtask] → [📋](./.task/IMPL-001.1.json)
-  - [x] **IMPL-001.2**: [Subtask] → [📋](./.task/IMPL-001.2.json) | [✅](./.summaries/IMPL-001.2-summary.md)
-
-- [x] **IMPL-002**: [Simple Task] → [📋](./.task/IMPL-002.json) | [✅](./.summaries/IMPL-002-summary.md)
-
-## Status Legend
-- `▸` = Container task (has subtasks)
-- `- [ ]` = Pending leaf task
-- `- [x]` = Completed leaf task
-- Maximum 2 levels: Main tasks and subtasks only
-```
-
-### 6.4. Output Files Diagram
-The command organizes outputs into a standard directory structure.
-```
-.workflow/{session-id}/
-├── IMPL_PLAN.md                     # Implementation plan
-├── TODO_LIST.md                     # Progress tracking
-├── .task/
-│   ├── IMPL-1.json                  # Container task
-│   ├── IMPL-1.1.json                # Leaf task with flow_control
-│   └── IMPL-1.2.json                # Leaf task with flow_control
-├── .brainstorming              # Input artifacts from brainstorm + synthesis
-│   ├── guidance-specification.md    # Finalized decisions (with resolved conflicts if any)
-│   └── {role}/analysis*.md          # Role analyses (enhanced by synthesis, with resolved conflicts if any)
-└── .process/
-    └── context-package.json         # Input from context-gather (smart context + conflict_risk status)
-```
-
-## 7. Artifact Integration
-The command intelligently detects and integrates artifacts from the `.brainstorming/` directory.
-
-#### Artifact Priority
-1.  **context-package.json** (critical): Primary source - smart context AND all brainstorm artifact paths in `brainstorm_artifacts` section + conflict_risk status
-2.  **role/analysis*.md** (highest): Paths from context-package.json → role-specific requirements, design specs, enhanced by synthesis, with resolved conflicts applied in-place
-3.  **guidance-specification.md** (high): Path from context-package.json → finalized decisions with resolved conflicts (if conflict_risk was >= medium)
-
-#### Artifact-Task Mapping
-Artifacts are mapped to tasks based on their relevance to the task's domain.
-- **Role analysis.md files**: Primary requirements source - all relevant role analyses included based on task type
-- **ui-designer/analysis.md**: Mapped to UI/Frontend tasks for design tokens, layouts, components
-- **system-architect/analysis.md**: Mapped to Architecture/Backend tasks for ADRs, APIs, patterns
-- **subject-matter-expert/analysis.md**: Mapped to tasks related to domain logic or standards
-- **data-architect/analysis.md**: Mapped to tasks involving data models, schemas, or APIs
-- **product-manager/analysis.md**: Mapped to all tasks for business requirements and user stories
-
-This ensures that each task has access to the most relevant and detailed specifications from role-specific analyses.
-
-## 8. CLI Execute Mode Details
-When using `--cli-execute`, each step in `implementation_approach` includes a `command` field with the execution command.
-
-**Key Points**:
-- **Sequential Steps**: Steps execute in order defined in `implementation_approach` array
-    - **Context Delivery**: Each codex command receives context via CONTEXT field: `@{context_package_path}` (role analyses loaded dynamically from context package)- **Multi-Step Tasks**: First step provides full context, subsequent steps use `resume --last` to maintain session continuity
-- **Step Dependencies**: Later steps reference outputs from earlier steps via `depends_on` field
-
-### Example 1: Agent Mode - Simple Task (Default, No Command)
-```json
-{
-  "id": "IMPL-001",
-  "title": "Implement user authentication module",
-  "context_package_path": ".workflow/WFS-session/.process/context-package.json",
-  "context": {
-    "depends_on": [],
-    "focus_paths": ["src/auth"],
-    "requirements": ["JWT-based authentication", "Login and registration endpoints"],
-    "acceptance": [
-      "JWT token generation working",
-      "Login and registration endpoints implemented",
-      "Tests passing with >70% coverage"
-    ]
-  },
-  "flow_control": {
-    "pre_analysis": [
-      {
-        "step": "load_role_analyses",
-        "action": "Load role analyses from context-package.json",
-        "commands": [
-          "Read({{context_package_path}})",
-          "Extract(brainstorm_artifacts.role_analyses[].files[].path)",
-          "Read(each extracted path)"
-        ],
-        "output_to": "role_analyses",
-        "on_error": "fail"
-      },
-      {
-        "step": "load_context",
-        "action": "Load context package for project structure",
-        "commands": ["Read({{context_package_path}})"],
-        "output_to": "context_pkg",
-        "on_error": "fail"
-      }
-    ],
-    "implementation_approach": [
-      {
-        "step": 1,
-        "title": "Implement JWT-based authentication",
-        "description": "Create authentication module using JWT following [role_analyses] requirements and [context_pkg] patterns",
-        "modification_points": [
-          "Create auth service with JWT generation",
-          "Implement login endpoint with credential validation",
-          "Implement registration endpoint with user creation",
-          "Add JWT middleware for route protection"
-        ],
-        "logic_flow": [
-          "User registers → validate input → hash password → create user",
-          "User logs in → validate credentials → generate JWT → return token",
-          "Protected routes → validate JWT → extract user → allow access"
-        ],
-        "depends_on": [],
-        "output": "auth_implementation"
-      }
-    ],
-    "target_files": ["src/auth/service.ts", "src/auth/middleware.ts", "src/routes/auth.ts"]
-  }
-}
-```
-
-### Example 2: CLI Execute Mode - Single Codex Step
-```json
-{
-  "id": "IMPL-002",
-  "title": "Implement user authentication module",
-  "context_package_path": ".workflow/WFS-session/.process/context-package.json",
-  "context": {
-    "depends_on": [],
-    "focus_paths": ["src/auth"],
-    "requirements": ["JWT-based authentication", "Login and registration endpoints"],
-    "acceptance": ["JWT generation working", "Endpoints implemented", "Tests passing"]
-  },
-  "flow_control": {
-    "pre_analysis": [
-      {
-        "step": "load_role_analyses",
-        "action": "Load role analyses from context-package.json",
-        "commands": [
-          "Read({{context_package_path}})",
-          "Extract(brainstorm_artifacts.role_analyses[].files[].path)",
-          "Read(each extracted path)"
-        ],
-        "output_to": "role_analyses",
-        "on_error": "fail"
-      }
-    ],
-    "implementation_approach": [
-      {
-        "step": 1,
-        "title": "Implement authentication with Codex",
-        "description": "Create JWT-based authentication module",
-        "command": "bash(codex -C src/auth --full-auto exec \"PURPOSE: Implement user authentication TASK: JWT-based auth with login/registration MODE: auto CONTEXT: @{{context_package_path}} EXPECTED: Complete auth module with tests RULES: Load role analyses from context-package.json → brainstorm_artifacts\" --skip-git-repo-check -s danger-full-access)",
-        "modification_points": ["Create auth service", "Implement endpoints", "Add JWT middleware"],
-        "logic_flow": ["Validate credentials", "Generate JWT", "Return token"],
-        "depends_on": [],
-        "output": "auth_implementation"
-      }
-    ],
-    "target_files": ["src/auth/service.ts", "src/auth/middleware.ts"]
-  }
-}
-```
-
-### Example 3: CLI Execute Mode - Multi-Step with Resume
-```json
-{
-  "id": "IMPL-003",
-  "title": "Implement role-based access control",
-  "context_package_path": ".workflow/WFS-session/.process/context-package.json",
-  "context": {
-    "depends_on": ["IMPL-002"],
-    "focus_paths": ["src/auth", "src/middleware"],
-    "requirements": ["User roles and permissions", "Route protection middleware"],
-    "acceptance": ["RBAC models created", "Middleware working", "Management API complete"]
-  },
-  "flow_control": {
-    "pre_analysis": [
-      {
-        "step": "load_context",
-        "action": "Load context and role analyses from context-package.json",
-        "commands": [
-          "Read({{context_package_path}})",
-          "Extract(brainstorm_artifacts.role_analyses[].files[].path)",
-          "Read(each extracted path)"
-        ],
-        "output_to": "full_context",
-        "on_error": "fail"
-      }
-    ],
-    "implementation_approach": [
-      {
-        "step": 1,
-        "title": "Create RBAC models",
-        "description": "Define role and permission data models",
-        "command": "bash(codex -C src/auth --full-auto exec \"PURPOSE: Create RBAC models TASK: Role and permission models MODE: auto CONTEXT: @{{context_package_path}} EXPECTED: Models with migrations RULES: Load role analyses from context-package.json → brainstorm_artifacts\" --skip-git-repo-check -s danger-full-access)",
-        "modification_points": ["Define role model", "Define permission model", "Create migrations"],
-        "logic_flow": ["Design schema", "Implement models", "Generate migrations"],
-        "depends_on": [],
-        "output": "rbac_models"
-      },
-      {
-        "step": 2,
-        "title": "Implement RBAC middleware",
-        "description": "Create route protection middleware using models from step 1",
-        "command": "bash(codex --full-auto exec \"PURPOSE: Create RBAC middleware TASK: Route protection middleware MODE: auto CONTEXT: RBAC models from step 1 EXPECTED: Middleware for route protection RULES: Use session patterns\" resume --last --skip-git-repo-check -s danger-full-access)",
-        "modification_points": ["Create permission checker", "Add route decorators", "Integrate with auth"],
-        "logic_flow": ["Check user role", "Validate permissions", "Allow/deny access"],
-        "depends_on": [1],
-        "output": "rbac_middleware"
-      },
-      {
-        "step": 3,
-        "title": "Add role management API",
-        "description": "Create CRUD endpoints for roles and permissions",
-        "command": "bash(codex --full-auto exec \"PURPOSE: Role management API TASK: CRUD endpoints for roles/permissions MODE: auto CONTEXT: Models and middleware from previous steps EXPECTED: Complete API with validation RULES: Maintain consistency\" resume --last --skip-git-repo-check -s danger-full-access)",
-        "modification_points": ["Create role endpoints", "Create permission endpoints", "Add validation"],
-        "logic_flow": ["Define routes", "Implement controllers", "Add authorization"],
-        "depends_on": [2],
-        "output": "role_management_api"
-      }
-    ],
-    "target_files": [
-      "src/models/Role.ts",
-      "src/models/Permission.ts",
-      "src/middleware/rbac.ts",
-      "src/routes/roles.ts"
-    ]
-  }
-}
-```
-
-**Pattern Summary**:
-- **Agent Mode (Example 1)**: No `command` field - agent executes via `modification_points` and `logic_flow`
-- **CLI Mode Single-Step (Example 2)**: One `command` field with full context package
-- **CLI Mode Multi-Step (Example 3)**: First step uses full context, subsequent steps use `resume --last`
-- **Context Delivery**: Context package provided via `@{...}` references in CONTEXT field
-
-## 9. Error Handling
-
-### Input Validation Errors
-| Error | Cause | Resolution |
-|-------|-------|------------|
-| Session not found | Invalid session ID | Verify session exists |
-| Context missing | Incomplete planning | Run context-gather first |
-| Invalid format | Corrupted results | Regenerate analysis |
-
-### Task Generation Errors
-| Error | Cause | Resolution |
-|-------|-------|------------|
-| Count exceeds limit | >10 tasks | Re-scope requirements |
-| Invalid structure | Missing fields | Fix analysis results |
-| Dependency cycle | Circular refs | Adjust dependencies |
-
-### Artifact Integration Errors
-| Error | Cause | Recovery |
-|-------|-------|----------|
-| Artifact not found | Missing output | Continue without artifacts |
-| Invalid format | Corrupted file | Skip artifact loading |
-| Path invalid | Moved/deleted | Update references |
-
-## 10. Integration & Usage
-
-### Command Chain
-- **Called By**: `/workflow:plan` (Phase 4)
-- **Calls**: None (terminal command)
-- **Followed By**: `/workflow:execute`, `/workflow:status`
-
-### Basic Usage
-```bash
-/workflow:tools:task-generate --session WFS-auth
-```
-
-## 11. Related Commands
-- `/workflow:plan` - Orchestrates entire planning
-- `/workflow:plan --cli-execute` - Planning with CLI execution mode
-- `/workflow:tools:context-gather` - Provides context package
-- `/workflow:tools:conflict-resolution` - Provides conflict resolution strategies (optional)
-- `/workflow:execute` - Executes generated tasks

.claude/commands/workflow/tools/tdd-coverage-analysis.md

@@ -1,6 +1,6 @@
 ---
 name: tdd-coverage-analysis
-description: Analyze test coverage and TDD cycle execution
+description: Analyze test coverage and TDD cycle execution with Red-Green-Refactor compliance verification
 argument-hint: "--session WFS-session-id"
 allowed-tools: Read(*), Write(*), Bash(*)
 ---
@@ -17,11 +17,43 @@ Analyze test coverage and verify Red-Green-Refactor cycle execution for TDD work
 - Verify TDD cycle execution (Red -> Green -> Refactor)
 - Generate coverage and cycle reports
 
+## Execution Process
+
+```
+Input Parsing:
+   ├─ Parse flags: --session
+   └─ Validation: session_id REQUIRED
+
+Phase 1: Extract Test Tasks
+   └─ Find TEST-*.json files and extract focus_paths
+
+Phase 2: Run Test Suite
+   └─ Decision (test framework):
+      ├─ Node.js → npm test --coverage --json
+      ├─ Python → pytest --cov --json-report
+      └─ Other → [test_command] --coverage --json
+
+Phase 3: Parse Coverage Data
+   ├─ Extract line coverage percentage
+   ├─ Extract branch coverage percentage
+   ├─ Extract function coverage percentage
+   └─ Identify uncovered lines/branches
+
+Phase 4: Verify TDD Cycle
+   └─ FOR each TDD chain (TEST-N.M → IMPL-N.M → REFACTOR-N.M):
+      ├─ Red Phase: Verify tests created and failed initially
+      ├─ Green Phase: Verify tests now pass
+      └─ Refactor Phase: Verify code quality improved
+
+Phase 5: Generate Analysis Report
+   └─ Create tdd-cycle-report.md with coverage metrics and cycle verification
+```
+
 ## Execution Lifecycle
 
 ### Phase 1: Extract Test Tasks
 ```bash
-find .workflow/{session_id}/.task/ -name 'TEST-*.json' -exec jq -r '.context.focus_paths[]' {} \;
+find .workflow/active/{session_id}/.task/ -name 'TEST-*.json' -exec jq -r '.context.focus_paths[]' {} \;
 ```
 
 **Output**: List of test directories/files from all TEST tasks
@@ -29,20 +61,20 @@ find .workflow/{session_id}/.task/ -name 'TEST-*.json' -exec jq -r '.context.foc
 ### Phase 2: Run Test Suite
 ```bash
 # Node.js/JavaScript
-npm test -- --coverage --json > .workflow/{session_id}/.process/test-results.json
+npm test -- --coverage --json > .workflow/active/{session_id}/.process/test-results.json
 
 # Python
-pytest --cov --json-report > .workflow/{session_id}/.process/test-results.json
+pytest --cov --json-report > .workflow/active/{session_id}/.process/test-results.json
 
 # Other frameworks (detect from project)
-[test_command] --coverage --json-output .workflow/{session_id}/.process/test-results.json
+[test_command] --coverage --json-output .workflow/active/{session_id}/.process/test-results.json
 ```
 
 **Output**: test-results.json with coverage data
 
 ### Phase 3: Parse Coverage Data
 ```bash
-jq '.coverage' .workflow/{session_id}/.process/test-results.json > .workflow/{session_id}/.process/coverage-report.json
+jq '.coverage' .workflow/active/{session_id}/.process/test-results.json > .workflow/active/{session_id}/.process/coverage-report.json
 ```
 
 **Extract**:
@@ -58,7 +90,7 @@ For each TDD chain (TEST-N.M -> IMPL-N.M -> REFACTOR-N.M):
 **1. Red Phase Verification**
 ```bash
 # Check TEST task summary
-cat .workflow/{session_id}/.summaries/TEST-N.M-summary.md
+cat .workflow/active/{session_id}/.summaries/TEST-N.M-summary.md
 ```
 
 Verify:
@@ -69,7 +101,7 @@ Verify:
 **2. Green Phase Verification**
 ```bash
 # Check IMPL task summary
-cat .workflow/{session_id}/.summaries/IMPL-N.M-summary.md
+cat .workflow/active/{session_id}/.summaries/IMPL-N.M-summary.md
 ```
 
 Verify:
@@ -80,7 +112,7 @@ Verify:
 **3. Refactor Phase Verification**
 ```bash
 # Check REFACTOR task summary
-cat .workflow/{session_id}/.summaries/REFACTOR-N.M-summary.md
+cat .workflow/active/{session_id}/.summaries/REFACTOR-N.M-summary.md
 ```
 
 Verify:
@@ -90,7 +122,7 @@ Verify:
 
 ### Phase 5: Generate Analysis Report
 
-Create `.workflow/{session_id}/.process/tdd-cycle-report.md`:
+Create `.workflow/active/{session_id}/.process/tdd-cycle-report.md`:
 
 ```markdown
 # TDD Cycle Analysis - {Session ID}
@@ -146,7 +178,7 @@ Create `.workflow/{session_id}/.process/tdd-cycle-report.md`:
 
 ## Output Files
 ```
-.workflow/{session-id}/
+.workflow/active//{session-id}/
 └── .process/
     ├── test-results.json         # Raw test execution results
     ├── coverage-report.json      # Parsed coverage data
@@ -272,10 +304,6 @@ Function Coverage: 91%
 
 Overall Compliance: 93/100
 
-Detailed report: .workflow/WFS-auth/.process/tdd-cycle-report.md
+Detailed report: .workflow/active/WFS-auth/.process/tdd-cycle-report.md
 ```
 
-## Related Commands
-- `/workflow:tdd-verify` - Uses this tool for verification
-- `/workflow:tools:task-generate-tdd` - Generates tasks this tool analyzes
-- `/workflow:execute` - Executes tasks before analysis

.claude/commands/workflow/tools/test-concept-enhanced.md

@@ -1,15 +1,15 @@
 ---
 name: test-concept-enhanced
-description: Analyze test requirements and generate test generation strategy using Gemini
+description: Coordinate test analysis workflow using cli-execution-agent to generate test strategy via Gemini
 argument-hint: "--session WFS-test-session-id --context path/to/test-context-package.json"
 examples:
-  - /workflow:tools:test-concept-enhanced --session WFS-test-auth --context .workflow/WFS-test-auth/.process/test-context-package.json
+  - /workflow:tools:test-concept-enhanced --session WFS-test-auth --context .workflow/active/WFS-test-auth/.process/test-context-package.json
 ---
 
 # Test Concept Enhanced Command
 
 ## Overview
-Specialized analysis tool for test generation workflows that uses Gemini to analyze test coverage gaps, implementation context, and generate comprehensive test generation strategies.
+Workflow coordinator that delegates test analysis to cli-execution-agent. Agent executes Gemini to analyze test coverage gaps, implementation context, and generate comprehensive test generation strategies.
 
 ## Core Philosophy
 - **Coverage-Driven**: Focus on identified test gaps from context analysis
@@ -19,18 +19,42 @@ Specialized analysis tool for test generation workflows that uses Gemini to anal
 - **No Code Generation**: Strategy and planning only, actual test generation happens in task execution
 
 ## Core Responsibilities
-- Parse test-context-package.json from test-context-gather
-- Analyze implementation summaries and coverage gaps
-- Study existing test patterns and conventions
-- Generate test generation strategy using Gemini
-- Produce TEST_ANALYSIS_RESULTS.md for task generation
+- Coordinate test analysis workflow using cli-execution-agent
+- Validate test-context-package.json prerequisites
+- Execute Gemini analysis via agent for test strategy generation
+- Validate agent outputs (gemini-test-analysis.md, TEST_ANALYSIS_RESULTS.md)
+
+## Execution Process
+
+```
+Input Parsing:
+   ├─ Parse flags: --session, --context
+   └─ Validation: Both REQUIRED
+
+Phase 1: Context Preparation (Command)
+   ├─ Load workflow-session.json
+   ├─ Verify test session type is "test-gen"
+   ├─ Validate test-context-package.json
+   └─ Determine strategy (Simple: 1-3 files | Medium: 4-6 | Complex: >6)
+
+Phase 2: Test Analysis Execution (Agent)
+   ├─ Execute Gemini analysis via cli-execution-agent
+   └─ Generate TEST_ANALYSIS_RESULTS.md
+
+Phase 3: Output Validation (Command)
+   ├─ Verify gemini-test-analysis.md exists
+   ├─ Validate TEST_ANALYSIS_RESULTS.md
+   └─ Confirm test requirements are actionable
+```
 
 ## Execution Lifecycle
 
-### Phase 1: Validation & Preparation
+### Phase 1: Context Preparation (Command Responsibility)
+
+**Command prepares session context and validates prerequisites.**
 
 1. **Session Validation**
-   - Load `.workflow/{test_session_id}/workflow-session.json`
+   - Load `.workflow/active/{test_session_id}/workflow-session.json`
    - Verify test session type is "test-gen"
    - Extract source session reference
 
@@ -40,428 +64,101 @@ Specialized analysis tool for test generation workflows that uses Gemini to anal
    - Extract coverage gaps and framework details
 
 3. **Strategy Determination**
-   - **Simple Test Generation** (1-3 files): Single Gemini analysis
-   - **Medium Test Generation** (4-6 files): Gemini comprehensive analysis
-   - **Complex Test Generation** (>6 files): Gemini analysis with modular approach
+   - **Simple** (1-3 files): Single Gemini analysis
+   - **Medium** (4-6 files): Comprehensive analysis
+   - **Complex** (>6 files): Modular analysis approach
 
-### Phase 2: Gemini Test Analysis
+### Phase 2: Test Analysis Execution (Agent Responsibility)
 
-**Tool Configuration**:
-```bash
-cd .workflow/{test_session_id}/.process && gemini -p "
-PURPOSE: Analyze test coverage gaps and design comprehensive test generation strategy
-TASK: Study implementation context, existing tests, and generate test requirements for missing coverage
-MODE: analysis
-CONTEXT: @{.workflow/{test_session_id}/.process/test-context-package.json}
+**Purpose**: Analyze test coverage gaps and generate comprehensive test strategy.
 
-**MANDATORY FIRST STEP**: Read and analyze test-context-package.json to understand:
-- Test coverage gaps from test_coverage.missing_tests[]
-- Implementation context from source_context.implementation_summaries[]
-- Existing test patterns from test_framework.conventions
-- Changed files requiring tests from source_context.implementation_summaries[].changed_files
+**Agent Invocation**:
+```javascript
+Task(
+  subagent_type="cli-execution-agent",
+  run_in_background=false,
+  description="Analyze test coverage gaps and generate test strategy",
+  prompt=`
+## TASK OBJECTIVE
+Analyze test requirements and generate comprehensive test generation strategy using Gemini CLI
 
-**ANALYSIS REQUIREMENTS**:
+## EXECUTION CONTEXT
+Session: {test_session_id}
+Source Session: {source_session_id}
+Working Dir: .workflow/active/{test_session_id}/.process
+Template: ~/.claude/workflows/cli-templates/prompts/test/test-concept-analysis.txt
 
-1. **Implementation Understanding**
-   - Load all implementation summaries from source session
-   - Understand implemented features, APIs, and business logic
-   - Extract key functions, classes, and modules
-   - Identify integration points and dependencies
+## EXECUTION STEPS
+1. Execute Gemini analysis:
+   ccw cli -p "$(cat ~/.claude/workflows/cli-templates/prompts/test/test-concept-analysis.txt)" --tool gemini --mode write --cd .workflow/active/{test_session_id}/.process
 
-2. **Existing Test Pattern Analysis**
-   - Study existing test files for patterns and conventions
-   - Identify test structure (describe/it, test suites, fixtures)
-   - Analyze assertion patterns and mocking strategies
-   - Extract test setup/teardown patterns
+2. Generate TEST_ANALYSIS_RESULTS.md:
+   Synthesize gemini-test-analysis.md into standardized format for task generation
+   Include: coverage assessment, test framework, test requirements, generation strategy, implementation targets
 
-3. **Coverage Gap Assessment**
-   - For each file in missing_tests[], analyze:
-     - File purpose and functionality
-     - Public APIs requiring test coverage
-     - Critical paths and edge cases
-     - Integration points requiring tests
-   - Prioritize tests: high (core logic), medium (utilities), low (helpers)
+## EXPECTED OUTPUTS
+1. gemini-test-analysis.md - Raw Gemini analysis
+2. TEST_ANALYSIS_RESULTS.md - Standardized test requirements document
 
-4. **Test Requirements Specification**
-   - For each missing test file, specify:
-     - **Test scope**: What needs to be tested
-     - **Test scenarios**: Happy path, error cases, edge cases, integration
-     - **Test data**: Required fixtures, mocks, test data
-     - **Dependencies**: External services, databases, APIs to mock
-     - **Coverage targets**: Functions/methods requiring tests
-
-5. **Test Generation Strategy**
-   - Determine test generation approach for each file
-   - Identify reusable test patterns from existing tests
-   - Plan test data and fixture requirements
-   - Define mocking strategy for dependencies
-   - Specify expected test file structure
-
-EXPECTED OUTPUT - Write to gemini-test-analysis.md:
-
-# Test Generation Analysis
-
-## 1. Implementation Context Summary
-- **Source Session**: {source_session_id}
-- **Implemented Features**: {feature_summary}
-- **Changed Files**: {list_of_implementation_files}
-- **Tech Stack**: {technologies_used}
-
-## 2. Test Coverage Assessment
-- **Existing Tests**: {count} files
-- **Missing Tests**: {count} files
-- **Coverage Percentage**: {percentage}%
-- **Priority Breakdown**:
-  - High Priority: {count} files (core business logic)
-  - Medium Priority: {count} files (utilities, helpers)
-  - Low Priority: {count} files (configuration, constants)
-
-## 3. Existing Test Pattern Analysis
-- **Test Framework**: {framework_name_and_version}
-- **File Naming Convention**: {pattern}
-- **Test Structure**: {describe_it_or_other}
-- **Assertion Style**: {expect_assert_should}
-- **Mocking Strategy**: {mocking_framework_and_patterns}
-- **Setup/Teardown**: {beforeEach_afterEach_patterns}
-- **Test Data**: {fixtures_factories_builders}
-
-## 4. Test Requirements by File
-
-### File: {implementation_file_path}
-**Test File**: {suggested_test_file_path}
-**Priority**: {high|medium|low}
-
-#### Scope
-- {description_of_what_needs_testing}
-
-#### Test Scenarios
-1. **Happy Path Tests**
-   - {scenario_1}
-   - {scenario_2}
-
-2. **Error Handling Tests**
-   - {error_scenario_1}
-   - {error_scenario_2}
-
-3. **Edge Case Tests**
-   - {edge_case_1}
-   - {edge_case_2}
-
-4. **Integration Tests** (if applicable)
-   - {integration_scenario_1}
-   - {integration_scenario_2}
-
-#### Test Data & Fixtures
-- {required_test_data}
-- {required_mocks}
-- {required_fixtures}
-
-#### Dependencies to Mock
-- {external_service_1}
-- {external_service_2}
-
-#### Coverage Targets
-- Function: {function_name} - {test_requirements}
-- Function: {function_name} - {test_requirements}
-
----
-[Repeat for each missing test file]
----
-
-## 5. Test Generation Strategy
-
-### Overall Approach
-- {strategy_description}
-
-### Test Generation Order
-1. {file_1} - {rationale}
-2. {file_2} - {rationale}
-3. {file_3} - {rationale}
-
-### Reusable Patterns
-- {pattern_1_from_existing_tests}
-- {pattern_2_from_existing_tests}
-
-### Test Data Strategy
-- {approach_to_test_data_and_fixtures}
-
-### Mocking Strategy
-- {approach_to_mocking_dependencies}
-
-### Quality Criteria
-- Code coverage target: {percentage}%
-- Test scenarios per function: {count}
-- Integration test coverage: {approach}
-
-## 6. Implementation Targets
-
-**Purpose**: Identify new test files to create
-
-**Format**: New test files only (no existing files to modify)
-
-**Test Files to Create**:
-1. **Target**: `tests/auth/TokenValidator.test.ts`
-   - **Type**: Create new test file
-   - **Purpose**: Test TokenValidator class
-   - **Scenarios**: 15 test cases covering validation logic, error handling, edge cases
-   - **Dependencies**: Mock JWT library, test fixtures for tokens
-
-2. **Target**: `tests/middleware/errorHandler.test.ts`
-   - **Type**: Create new test file
-   - **Purpose**: Test error handling middleware
-   - **Scenarios**: 8 test cases for different error types and response formats
-   - **Dependencies**: Mock Express req/res/next, error fixtures
-
-[List all test files to create]
-
-## 7. Success Metrics
-- **Test Coverage Goal**: {target_percentage}%
-- **Test Quality**: All scenarios covered (happy, error, edge, integration)
-- **Convention Compliance**: Follow existing test patterns
-- **Maintainability**: Clear test descriptions, reusable fixtures
-
-RULES:
-- Focus on TEST REQUIREMENTS and GENERATION STRATEGY, NOT code generation
-- Study existing test patterns thoroughly for consistency
-- Prioritize critical business logic tests
-- Specify clear test scenarios and coverage targets
-- Identify all dependencies requiring mocks
-- **MUST write output to .workflow/{test_session_id}/.process/gemini-test-analysis.md**
-- Do NOT generate actual test code or implementation
-- Output ONLY test analysis and generation strategy
-" --approval-mode yolo
+## QUALITY VALIDATION
+- Both output files exist and are complete
+- All required sections present in TEST_ANALYSIS_RESULTS.md
+- Test requirements are actionable and quantified
+- Test scenarios cover happy path, errors, edge cases
+- Dependencies and mocks clearly identified
+`
+)
 ```
 
-**Output Location**: `.workflow/{test_session_id}/.process/gemini-test-analysis.md`
+**Output Files**:
+- `.workflow/active/{test_session_id}/.process/gemini-test-analysis.md`
+- `.workflow/active/{test_session_id}/.process/TEST_ANALYSIS_RESULTS.md`
 
-### Phase 3: Results Synthesis
+### Phase 3: Output Validation (Command Responsibility)
 
-1. **Output Validation**
-   - Verify `gemini-test-analysis.md` exists and is complete
-   - Validate all required sections present
-   - Check test requirements are actionable
+**Command validates agent outputs.**
 
-2. **Quality Assessment**
-   - Test scenarios cover happy path, errors, edge cases
-   - Dependencies and mocks clearly identified
-   - Test generation strategy is practical
-   - Coverage targets are reasonable
-
-### Phase 4: TEST_ANALYSIS_RESULTS.md Generation
-
-Synthesize Gemini analysis into standardized format:
-
-```markdown
-# Test Generation Analysis Results
-
-## Executive Summary
-- **Test Session**: {test_session_id}
-- **Source Session**: {source_session_id}
-- **Analysis Timestamp**: {timestamp}
-- **Coverage Gap**: {missing_test_count} files require tests
-- **Test Framework**: {framework}
-- **Overall Strategy**: {high_level_approach}
-
----
-
-## 1. Coverage Assessment
-
-### Current Coverage
-- **Existing Tests**: {count} files
-- **Implementation Files**: {count} files
-- **Coverage Percentage**: {percentage}%
-
-### Missing Tests (Priority Order)
-1. **High Priority** ({count} files)
-   - {file_1} - {reason}
-   - {file_2} - {reason}
-
-2. **Medium Priority** ({count} files)
-   - {file_1} - {reason}
-
-3. **Low Priority** ({count} files)
-   - {file_1} - {reason}
-
----
-
-## 2. Test Framework & Conventions
-
-### Framework Configuration
-- **Framework**: {framework_name}
-- **Version**: {version}
-- **Test Pattern**: {file_pattern}
-- **Test Directory**: {directory_structure}
-
-### Conventions
-- **File Naming**: {convention}
-- **Test Structure**: {describe_it_blocks}
-- **Assertions**: {assertion_library}
-- **Mocking**: {mocking_framework}
-- **Setup/Teardown**: {beforeEach_afterEach}
-
-### Example Pattern (from existing tests)
-```
-{example_test_structure_from_analysis}
-```
-
----
-
-## 3. Test Requirements by File
-
-[For each missing test, include:]
-
-### Test File: {test_file_path}
-**Implementation**: {implementation_file}
-**Priority**: {high|medium|low}
-**Estimated Test Count**: {count}
-
-#### Test Scenarios
-1. **Happy Path**: {scenarios}
-2. **Error Handling**: {scenarios}
-3. **Edge Cases**: {scenarios}
-4. **Integration**: {scenarios}
-
-#### Dependencies & Mocks
-- {dependency_1_to_mock}
-- {dependency_2_to_mock}
-
-#### Test Data Requirements
-- {fixture_1}
-- {fixture_2}
-
----
-
-## 4. Test Generation Strategy
-
-### Generation Approach
-{overall_strategy_description}
-
-### Generation Order
-1. {test_file_1} - {rationale}
-2. {test_file_2} - {rationale}
-3. {test_file_3} - {rationale}
-
-### Reusable Components
-- **Test Fixtures**: {common_fixtures}
-- **Mock Patterns**: {common_mocks}
-- **Helper Functions**: {test_helpers}
-
-### Quality Targets
-- **Coverage Goal**: {percentage}%
-- **Scenarios per Function**: {min_count}
-- **Integration Coverage**: {approach}
-
----
-
-## 5. Implementation Targets
-
-**Purpose**: New test files to create (code-developer will generate these)
-
-**Test Files to Create**:
-
-1. **Target**: `tests/auth/TokenValidator.test.ts`
-   - **Implementation Source**: `src/auth/TokenValidator.ts`
-   - **Test Scenarios**: 15 (validation, error handling, edge cases)
-   - **Dependencies**: Mock JWT library, token fixtures
-   - **Priority**: High
-
-2. **Target**: `tests/middleware/errorHandler.test.ts`
-   - **Implementation Source**: `src/middleware/errorHandler.ts`
-   - **Test Scenarios**: 8 (error types, response formats)
-   - **Dependencies**: Mock Express, error fixtures
-   - **Priority**: High
-
-[List all test files with full specifications]
-
----
-
-## 6. Success Criteria
-
-### Coverage Metrics
-- Achieve {target_percentage}% code coverage
-- All public APIs have tests
-- Critical paths fully covered
-
-### Quality Standards
-- All test scenarios covered (happy, error, edge, integration)
-- Follow existing test conventions
-- Clear test descriptions and assertions
-- Maintainable test structure
-
-### Validation Approach
-- Run full test suite after generation
-- Verify coverage with coverage tool
-- Manual review of test quality
-- Integration test validation
-
----
-
-## 7. Reference Information
-
-### Source Context
-- **Implementation Summaries**: {paths}
-- **Existing Tests**: {example_tests}
-- **Documentation**: {relevant_docs}
-
-### Analysis Tools
-- **Gemini Analysis**: gemini-test-analysis.md
-- **Coverage Tools**: {coverage_tool_if_detected}
-```
-
-**Output Location**: `.workflow/{test_session_id}/.process/TEST_ANALYSIS_RESULTS.md`
+- Verify `gemini-test-analysis.md` exists and is complete
+- Validate `TEST_ANALYSIS_RESULTS.md` generated by agent
+- Check required sections present
+- Confirm test requirements are actionable
 
 ## Error Handling
 
 ### Validation Errors
-| Error | Cause | Resolution |
-|-------|-------|------------|
-| Missing context package | test-context-gather not run | Run test-context-gather first |
-| No coverage gaps | All files have tests | Skip test generation, proceed to test execution |
-| No test framework detected | Missing test dependencies | Request user to configure test framework |
-| Invalid source session | Source session incomplete | Complete implementation first |
+| Error | Resolution |
+|-------|------------|
+| Missing context package | Run test-context-gather first |
+| No coverage gaps | Skip test generation, proceed to execution |
+| No test framework detected | Configure test framework |
+| Invalid source session | Complete implementation first |
 
-### Gemini Execution Errors
-| Error | Cause | Recovery |
-|-------|-------|----------|
-| Timeout | Large project analysis | Reduce scope, analyze by module |
-| Output incomplete | Token limit exceeded | Retry with focused analysis |
-| No output file | Write permission error | Check directory permissions |
+### Execution Errors
+| Error | Recovery |
+|-------|----------|
+| Gemini timeout | Reduce scope, analyze by module |
+| Output incomplete | Retry with focused analysis |
+| No output file | Check directory permissions |
 
-### Fallback Strategy
-- If Gemini fails, generate basic TEST_ANALYSIS_RESULTS.md from context package
-- Use coverage gaps and framework info to create minimal requirements
-- Provide guidance for manual test planning
+**Fallback Strategy**: Generate basic TEST_ANALYSIS_RESULTS.md from context package if Gemini fails
 
-## Performance Optimization
+## Integration & Usage
 
-- **Focused Analysis**: Only analyze files with missing tests
-- **Pattern Reuse**: Study existing tests for quick pattern extraction
-- **Parallel Operations**: Load implementation summaries in parallel
-- **Timeout Management**: 20-minute limit for Gemini analysis
+### Command Chain
+- **Called By**: `/workflow:test-gen` (Phase 4: Analysis)
+- **Requires**: `test-context-package.json` from `/workflow:tools:test-context-gather`
+- **Followed By**: `/workflow:tools:test-task-generate`
 
-## Integration
+### Performance
+- Focused analysis: Only analyze files with missing tests
+- Pattern reuse: Study existing tests for quick extraction
+- Timeout: 20-minute limit for analysis
 
-### Called By
-- `/workflow:test-gen` (Phase 4: Analysis)
+### Success Criteria
+- Valid TEST_ANALYSIS_RESULTS.md generated
+- All missing tests documented with actionable requirements
+- Test scenarios cover happy path, errors, edge cases, integration
+- Dependencies and mocks clearly identified
+- Test generation strategy is practical
+- Output follows existing test conventions
 
-### Requires
-- `/workflow:tools:test-context-gather` output (test-context-package.json)
-
-### Followed By
-- `/workflow:tools:test-task-generate` - Generates test task JSON with code-developer invocation
-
-## Success Criteria
-
-- ✅ Valid TEST_ANALYSIS_RESULTS.md generated
-- ✅ All missing tests documented with requirements
-- ✅ Test scenarios cover happy path, errors, edge cases
-- ✅ Dependencies and mocks identified
-- ✅ Test generation strategy is actionable
-- ✅ Execution time < 20 minutes
-- ✅ Output follows existing test conventions
-
-## Related Commands
-
-- `/workflow:tools:test-context-gather` - Provides input context
-- `/workflow:tools:test-task-generate` - Consumes analysis results
-- `/workflow:test-gen` - Main test generation workflow