Enhance search functionality and indexing pipeline

- Updated `cmd_search` to include line numbers and content in search results. - Modified `IndexingPipeline` to handle start and end line numbers for chunks. - Enhanced `FTSEngine` to support storing line metadata in the database. - Improved `SearchPipeline` to return line numbers and full content in search results. - Added unit tests for bridge, FTS delete operations, metadata store, and watcher functionality. - Introduced a `.gitignore` file to exclude specific directories.
2026-03-18 18:48:48 +08:00 · 2026-03-17 14:55:27 +08:00
parent bfe5426b7e
commit 0f02b75be1
25 changed files with 2014 additions and 1482 deletions
--- a/.claude/skills/prompt-generator/SKILL.md
+++ b/.claude/skills/prompt-generator/SKILL.md
@@ -1,19 +1,20 @@
 ---
 name: prompt-generator
-description: Generate or convert Claude Code prompt files — command orchestrators, agent role definitions, or style conversion of existing files. Follows GSD-style content separation with built-in quality gates. Triggers on "create command", "new command", "create agent", "new agent", "convert command", "convert agent", "prompt generator".
+description: Generate or convert Claude Code prompt files — command orchestrators, skill files, agent role definitions, or style conversion of existing files. Follows GSD-style content separation with built-in quality gates. Triggers on "create command", "new command", "create skill", "new skill", "create agent", "new agent", "convert command", "convert skill", "convert agent", "prompt generator", "优化".
 allowed-tools: Read, Write, Edit, Bash, Glob, AskUserQuestion
 ---

 <purpose>
-Generate or convert Claude Code prompt files with concrete, domain-specific content. Three modes:
+Generate or convert Claude Code prompt files with concrete, domain-specific content. Four modes:

 - **Create command** — new orchestration workflow at `.claude/commands/` or `~/.claude/commands/`
+- **Create skill** — new skill file at `.claude/skills/*/SKILL.md` (progressive loading, no @ refs)
 - **Create agent** — new role + expertise file at `.claude/agents/`
- **Convert** — restyle existing command/agent to GSD conventions with zero content loss
+- **Convert** — restyle existing command/skill/agent to GSD conventions with zero content loss

-Content separation principle (from GSD): commands own orchestration flow; agents own domain knowledge.
+Content separation principle (from GSD): commands/skills own orchestration flow; agents own domain knowledge. Skills are a variant of commands but loaded progressively inline — they CANNOT use `@` file references.

-Invoked when user requests "create command", "new command", "create agent", "new agent", "convert command", "convert agent", or "prompt generator".
+Invoked when user requests "create command", "new command", "create skill", "new skill", "create agent", "new agent", "convert command", "convert skill", "convert agent", "prompt generator", or "优化".
 </purpose>

 <required_reading>
@@ -33,11 +34,17 @@ Parse `$ARGUMENTS` to determine what to generate.
 | Signal | Type |
 |--------|------|
 | "command", "workflow", "orchestrator" in args | `command` |
+| "skill", "SKILL.md" in args, or path contains `.claude/skills/` | `skill` |
 | "agent", "role", "worker" in args | `agent` |
-| "convert", "restyle", "refactor" + file path in args | `convert` |
+| "convert", "restyle", "refactor", "optimize", "优化" + file path in args | `convert` |
 | Ambiguous or missing | Ask user |

-**Convert mode detection:** If args contain a file path (`.md` extension) + conversion keywords, enter convert mode. Extract `$SOURCE_PATH` from args.
+**Convert mode detection:** If args contain a file path (`.md` extension) + conversion keywords, enter convert mode. Extract `$SOURCE_PATH` from args. Auto-detect source type from path:
+- `.claude/commands/` → command
+- `.claude/skills/*/SKILL.md` → skill
+- `.claude/agents/` → agent
+
+**Skill vs Command distinction:** Skills (`.claude/skills/*/SKILL.md`) are loaded **progressively inline** into the conversation context. They CANNOT use `@` file references — only `Read()` tool calls within process steps. See `@specs/command-design-spec.md` → "Skill Variant" section.

 If ambiguous:

@@ -47,13 +54,14 @@ AskUserQuestion(
  question: "What type of prompt file do you want to generate?",
  options: [
    { label: "Command", description: "New orchestration workflow — process steps, user interaction, agent spawning" },
+    { label: "Skill", description: "New skill file — progressive loading, no @ refs, inline Read() for external files" },
    { label: "Agent", description: "New role definition — identity, domain expertise, behavioral rules" },
-    { label: "Convert", description: "Restyle existing command/agent to GSD conventions (zero content loss)" }
+    { label: "Convert", description: "Restyle existing command/agent/skill to GSD conventions (zero content loss)" }
  ]
 )
 ```

-Store as `$ARTIFACT_TYPE` (`command` | `agent` | `convert`).
+Store as `$ARTIFACT_TYPE` (`command` | `skill` | `agent` | `convert`).

 ## 2. Validate Parameters

@@ -101,6 +109,12 @@ Else:
  $TARGET_PATH = {base}/{$NAME}.md
 ```

+**Skill:**
+
+```
+$TARGET_PATH = .claude/skills/{$NAME}/SKILL.md
+```
+
 **Agent:**

 ```
@@ -179,6 +193,31 @@ Generate a complete command file with:
 - Shell blocks use heredoc for multi-line, quote all variables
 - Include `<auto_mode>` section if command supports `--auto` flag

+### 5a-skill. Skill Generation (variant of command)
+
+Follow `@specs/command-design-spec.md` → "Skill Variant" section.
+
+Skills are command-like orchestrators but loaded **progressively inline** — they CANNOT use `@` file references.
+
+Generate a complete skill file with:
+
+1. **`<purpose>`** — 2-3 sentences: what + when + what it produces
+2. **NO `<required_reading>`** — skills cannot use `@` refs. External files loaded via `Read()` within process steps.
+3. **`<process>`** — numbered steps (GSD workflow style):
+   - Step 1: Initialize / parse arguments / set workflow preferences
+   - Steps 2-N: Domain-specific orchestration logic with inline `Read("phases/...")` for phase files
+   - Each step: validation, agent spawning via `Agent()`, error handling
+   - Final step: completion status or handoff to next skill via `Skill()`
+4. **`<success_criteria>`** — checkbox list of verifiable conditions
+
+**Skill-specific writing rules:**
+- **NO `<required_reading>` tag** — `@` syntax not supported in skills
+- **NO `@path` references** anywhere in the file — use `Read("path")` within `<process>` steps
+- Phase files loaded on-demand: `Read("phases/01-xxx.md")` within the step that needs it
+- Frontmatter uses `allowed-tools:` (not `argument-hint:`)
+- `<offer_next>` is optional — skills often chain via `Skill()` calls
+- `<auto_mode>` can be inline within `<process>` step 1 or as standalone section
+
 ### 5b. Agent Generation

 Follow `@specs/agent-design-spec.md` and `@templates/agent-md.md`.
@@ -225,11 +264,20 @@ $INVENTORY = {

 | Signal | Type |
 |--------|------|
+| Path in `.claude/skills/*/SKILL.md` | skill |
+| `allowed-tools:` in frontmatter + path in `.claude/skills/` | skill |
 | Contains `<process>`, `<step>`, numbered `## N.` steps | command |
 | Contains `<role>`, `tools:` in frontmatter, domain sections | agent |
-| Flat markdown with `## Implementation`, `## Phase N` | command (unstructured) |
+| Flat markdown with `## Implementation`, `## Phase N` + in skills dir | skill (unstructured) |
+| Flat markdown with `## Implementation`, `## Phase N` + in commands dir | command (unstructured) |
 | Flat prose with role description, no process steps | agent (unstructured) |

+**Skill-specific conversion rules:**
+- **NO `<required_reading>`** — skills cannot use `@` file references (progressive loading)
+- **NO `@path` references** anywhere — replace with `Read("path")` within `<process>` steps
+- If source has `@specs/...` or `@phases/...` refs, convert to `Read("specs/...")` / `Read("phases/...")`
+- Follow `@specs/conversion-spec.md` → "Skill Conversion Rules" section
+
 **Step 5c.3: Build conversion map.**

 Map every source section to its target location. Follow `@specs/conversion-spec.md` transformation rules.
@@ -293,6 +341,20 @@ Set `$TARGET_PATH = $SOURCE_PATH` (in-place conversion) unless user specifies ou
 | `<success_criteria>` | 4+ checkbox items, all verifiable |
 | Content separation | No domain expertise embedded — only orchestration |

+### 6b-skill. Skill-Specific Checks
+
+| Check | Pass Condition |
+|-------|---------------|
+| `<purpose>` | 2-3 sentences, no placeholders |
+| **NO `<required_reading>`** | Must NOT contain `<required_reading>` tag |
+| **NO `@` file references** | Zero `@specs/`, `@phases/`, `@./` patterns in prose |
+| `<process>` with numbered steps | At least 3 `## N.` headers |
+| Step 1 is initialization | Parses args, sets workflow preferences |
+| Phase file loading | Uses `Read("phases/...")` within process steps (if has phases) |
+| `<success_criteria>` | 4+ checkbox items, all verifiable |
+| Frontmatter `allowed-tools` | Present and lists required tools |
+| Content separation | No domain expertise embedded — only orchestration |
+
 ### 6c. Agent-Specific Checks

 | Check | Pass Condition |
--- a/.claude/skills/prompt-generator/specs/command-design-spec.md
+++ b/.claude/skills/prompt-generator/specs/command-design-spec.md
@@ -36,6 +36,7 @@ allowed-tools: Tool1, Tool2  # Optional: restricted tool set
 .claude/commands/deploy.md           # Top-level command
 .claude/commands/issue/create.md     # Grouped command
 ~/.claude/commands/global-status.md  # User-level command
+.claude/skills/my-skill/SKILL.md     # Skill file (see Skill Variant below)
 ```

 ## Content Structure
@@ -45,12 +46,60 @@ Commands use XML semantic tags with process steps inside `<process>`:
 | Tag | Required | Purpose |
 |-----|----------|---------|
 | `<purpose>` | Yes | What + when + what it produces (2-3 sentences) |
-| `<required_reading>` | Yes | @ file references loaded before execution |
+| `<required_reading>` | Commands only | @ file references loaded before execution |
 | `<process>` | Yes | Steps — numbered or named (see Step Styles below) |
 | `<auto_mode>` | Optional | Behavior when `--auto` flag present |
 | `<offer_next>` | Recommended | Formatted completion status + next actions |
 | `<success_criteria>` | Yes | Checkbox list of verifiable conditions |

+## Skill Variant
+
+Skills (`.claude/skills/*/SKILL.md`) follow command structure with critical differences due to **progressive loading** — skills are loaded inline into the conversation context, NOT via file resolution.
+
+### Key Differences: Skill vs Command
+
+| Aspect | Command | Skill |
+|--------|---------|-------|
+| Location | `.claude/commands/` | `.claude/skills/*/SKILL.md` |
+| Loading | Slash-command invocation, `@` refs resolved | Progressive inline loading into conversation |
+| `<required_reading>` | Yes — `@path` refs auto-resolved | **NO** — `@` refs do NOT work in skills |
+| External file access | `@` references | `Read()` tool calls within `<process>` steps |
+| Phase files | N/A | `Read("phases/01-xxx.md")` within process steps |
+| Frontmatter | `name`, `description`, `argument-hint` | `name`, `description`, `allowed-tools` |
+
+### Skill-Specific Rules
+
+1. **NO `<required_reading>` tag** — Skills cannot use `@` file references. All external context must be loaded via `Read()` tool calls within `<process>` steps.
+
+2. **Progressive phase loading** — For multi-phase skills with phase files in `phases/` subdirectory, use inline `Read()`:
+   ```javascript
+   // Within process step: Load phase doc on-demand
+   Read("phases/01-session-discovery.md")
+   // Execute phase logic...
+   ```
+
+3. **Self-contained content** — All instructions, rules, and logic must be directly in the SKILL.md or loaded via `Read()` at runtime. No implicit file dependencies.
+
+4. **Frontmatter uses `allowed-tools:`** instead of `argument-hint:`:
+   ```yaml
+   ---
+   name: my-skill
+   description: What this skill does
+   allowed-tools: Agent, Read, Write, Bash, Glob, Grep
+   ---
+   ```
+
+### Skill Content Structure
+
+| Tag | Required | Purpose |
+|-----|----------|---------|
+| `<purpose>` | Yes | What + when + what it produces (2-3 sentences) |
+| `<process>` | Yes | Steps with inline `Read()` for external files |
+| `<auto_mode>` | Optional | Behavior when `-y`/`--yes` flag present |
+| `<success_criteria>` | Yes | Checkbox list of verifiable conditions |
+
+**Note**: `<offer_next>` is less common in skills since skills often chain to other skills via `Skill()` calls.
+
 ## Step Styles

 GSD uses two step styles. Choose based on command nature:
--- a/.claude/skills/prompt-generator/specs/conversion-spec.md
+++ b/.claude/skills/prompt-generator/specs/conversion-spec.md
@@ -36,6 +36,62 @@ Conversion Summary:
  New sections added: {list of TODO sections}
 ```

+## Artifact Type Detection
+
+Before applying conversion rules, determine the source type:
+
+| Source Location | Type |
+|----------------|------|
+| `.claude/commands/**/*.md` | command |
+| `.claude/skills/*/SKILL.md` | skill |
+| `.claude/agents/*.md` | agent |
+
+**Skill detection signals**: `allowed-tools:` in frontmatter, located in `.claude/skills/` directory, progressive phase loading pattern (`Read("phases/...")`)
+
+## Skill Conversion Rules
+
+### Critical: No @ References
+
+Skills are loaded **progressively inline** into the conversation context. They CANNOT use `@` file references — these only work in commands.
+
+### Source Pattern → Target Pattern (Skill)
+
+| Source Style | Target Style |
+|-------------|-------------|
+| `# Title` + flat markdown overview | `<purpose>` (2-3 sentences) |
+| `## Implementation` / `## Execution Flow` / `## Phase Summary` | `<process>` with numbered `## N.` steps |
+| Phase file references as prose | `Read("phases/...")` calls within process steps |
+| `## Success Criteria` / `## Coordinator Checklist` | `<success_criteria>` with checkbox list |
+| `## Auto Mode` / `## Auto Mode Defaults` | `<auto_mode>` section |
+| `## Error Handling` | Preserve as-is within `<process>` or as standalone section |
+| Code blocks, tables, ASCII diagrams | **Preserve exactly** |
+
+### What NOT to Add (Skill-Specific)
+
+| Element | Why NOT |
+|---------|---------|
+| `<required_reading>` | Skills cannot use `@` refs — progressive loading |
+| `@specs/...` or `@phases/...` | `@` syntax not supported in skills |
+| `<offer_next>` | Skills chain via `Skill()` calls, not offer menus |
+
+### What to ADD (Skill-Specific)
+
+| Missing Element | Add |
+|----------------|-----|
+| `<purpose>` | Extract from overview/description |
+| `<process>` wrapper | Wrap implementation steps |
+| `<success_criteria>` | Generate from coordinator checklist or existing content |
+| `<auto_mode>` | If auto mode behavior exists, wrap in tag |
+
+### Frontmatter Conversion (Skill)
+
+| Source Field | Target Field | Transformation |
+|-------------|-------------|----------------|
+| `name` | `name` | Keep as-is |
+| `description` | `description` | Keep as-is |
+| `allowed-tools` | `allowed-tools` | Keep as-is |
+| Missing `allowed-tools` | `allowed-tools` | Infer from content |
+
 ## Command Conversion Rules

 ### Source Pattern → Target Pattern
--- a/.claude/skills/workflow-lite-plan/SKILL.md
+++ b/.claude/skills/workflow-lite-plan/SKILL.md
@@ -4,17 +4,18 @@ description: Lightweight planning skill - task analysis, multi-angle exploration
 allowed-tools: Skill, Agent, AskUserQuestion, TodoWrite, Read, Write, Edit, Bash, Glob, Grep
 ---

-# Workflow-Lite-Plan
-
+<purpose>
 Planning pipeline: explore → clarify → plan → confirm → handoff to lite-execute.
+Produces exploration results, a structured plan (plan.json), independent task files (.task/TASK-*.json), and hands off to lite-execute for implementation.
+</purpose>

---
+<process>

-## Context Isolation
+## 1. Context Isolation

 > **CRITICAL**: If invoked from analyze-with-file (via "执行任务"), the analyze-with-file session is **COMPLETE** and all its phase instructions are FINISHED and MUST NOT be referenced. Only follow LP-Phase 1-5 defined in THIS document. Phase numbers are INDEPENDENT of any prior workflow.

-## Input
+## 2. Input

 ```
 <task-description>         Task description or path to .md file (required)
@@ -27,7 +28,7 @@ Planning pipeline: explore → clarify → plan → confirm → handoff to lite-

 **Note**: Workflow preferences (`autoYes`, `forceExplore`) must be initialized at skill start. If not provided by caller, skill will prompt user for workflow mode selection.

-## Output Artifacts
+## 3. Output Artifacts

 | Artifact | Description |
 |----------|-------------|
@@ -43,14 +44,7 @@ Planning pipeline: explore → clarify → plan → confirm → handoff to lite-

 **Schema Reference**: `~/.ccw/workflows/cli-templates/schemas/plan-overview-base-schema.json`

-## Auto Mode Defaults
-
-When `workflowPreferences.autoYes === true` (entire plan+execute workflow):
- **Clarification**: Skipped | **Plan Confirmation**: Allow & Execute | **Execution**: Auto | **Review**: Skip
-
-Auto mode authorizes the complete plan-and-execute workflow with a single confirmation. No further prompts.
-
-## Phase Summary
+## 4. Phase Summary

 | Phase | Core Action | Output |
 |-------|-------------|--------|
@@ -61,9 +55,7 @@ Auto mode authorizes the complete plan-and-execute workflow with a single confir
 | LP-4 | Display plan → AskUserQuestion (Confirm + Execution + Review) | userSelection |
 | LP-5 | Build executionContext → Skill("lite-execute") | handoff (Mode 1) |

-## Implementation
-
-### LP-Phase 0: Workflow Preferences Initialization
+## 5. LP-Phase 0: Workflow Preferences Initialization

 ```javascript
 if (typeof workflowPreferences === 'undefined' || workflowPreferences === null) {
@@ -74,7 +66,7 @@ if (typeof workflowPreferences === 'undefined' || workflowPreferences === null)
 }
 ```

-### LP-Phase 1: Intelligent Multi-Angle Exploration
+## 6. LP-Phase 1: Intelligent Multi-Angle Exploration

 **Session Setup** (MANDATORY):
 ```javascript
@@ -248,9 +240,7 @@ console.log(`Exploration complete: ${explorationManifest.explorations.map(e => e

 **Output**: `exploration-{angle}.json` (1-4 files) + `explorations-manifest.json`

---
-
-### LP-Phase 2: Clarification (Optional, Multi-Round)
+## 7. LP-Phase 2: Clarification (Optional, Multi-Round)

 **Skip if**: No exploration or `clarification_needs` is empty across all explorations

@@ -307,9 +297,7 @@ if (workflowPreferences.autoYes) {

 **Output**: `clarificationContext` (in-memory)

---
-
-### LP-Phase 3: Planning
+## 8. LP-Phase 3: Planning

 **IMPORTANT**: LP-Phase 3 is **planning only** — NO code execution. All execution happens in LP-Phase 5 via lite-execute.

@@ -431,9 +419,7 @@ ${complexity}

 // TodoWrite: Phase 3 → completed, Phase 4 → in_progress

---
-
-### LP-Phase 4: Task Confirmation & Execution Selection
+## 9. LP-Phase 4: Task Confirmation & Execution Selection

 **Display Plan**:
 ```javascript
@@ -499,9 +485,7 @@ if (workflowPreferences.autoYes) {

 // TodoWrite: Phase 4 → completed `[${userSelection.execution_method} + ${userSelection.code_review_tool}]`, Phase 5 → in_progress

---
-
-### LP-Phase 5: Handoff to Execution
+## 10. LP-Phase 5: Handoff to Execution

 **CRITICAL**: lite-plan NEVER executes code directly. ALL execution goes through lite-execute.

@@ -562,7 +546,7 @@ Skill("lite-execute")
 // executionContext passed as global variable (Mode 1: In-Memory Plan)
 ```

-## Session Folder Structure
+## 11. Session Folder Structure

 ```
 .workflow/.lite-plan/{task-slug}-{YYYY-MM-DD}/
@@ -576,7 +560,7 @@ Skill("lite-execute")
    └── ...
 ```

-## Error Handling
+## 12. Error Handling

 | Error | Resolution |
 |-------|------------|
@@ -585,3 +569,26 @@ Skill("lite-execute")
 | 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 |
+
+</process>
+
+<auto_mode>
+When `workflowPreferences.autoYes === true` (entire plan+execute workflow):
+- **Clarification**: Skipped | **Plan Confirmation**: Allow & Execute | **Execution**: Auto | **Review**: Skip
+
+Auto mode authorizes the complete plan-and-execute workflow with a single confirmation. No further prompts.
+</auto_mode>
+
+<success_criteria>
+- [ ] Workflow preferences (autoYes, forceExplore) initialized at LP-Phase 0
+- [ ] Complexity assessed and exploration angles selected appropriately
+- [ ] Parallel exploration agents launched with run_in_background=false
+- [ ] Explorations manifest built from auto-discovered files
+- [ ] Clarification needs aggregated, deduped, and presented in batches of 4
+- [ ] Plan generated via direct Claude (Low) or cli-lite-planning-agent (Medium/High)
+- [ ] Plan output as two-layer: plan.json (task_ids[]) + .task/TASK-*.json
+- [ ] User confirmation collected (or auto-approved in auto mode)
+- [ ] executionContext fully built with all artifacts and session references
+- [ ] Handoff to lite-execute via Skill("lite-execute") with executionContext
+- [ ] No code execution in planning phases -- all execution deferred to lite-execute
+</success_criteria>
--- a/.claude/skills/workflow-plan/SKILL.md
+++ b/.claude/skills/workflow-plan/SKILL.md
@@ -4,18 +4,20 @@ description: Unified planning skill - 4-phase planning workflow, plan verificati
 allowed-tools: Skill, Agent, AskUserQuestion, TodoWrite, Read, Write, Edit, Bash, Glob, Grep
 ---

-# Workflow Plan
+<purpose>
+Unified planning skill combining 4-phase planning workflow, plan quality verification, and interactive replanning. Produces IMPL_PLAN.md, task JSONs, verification reports, and manages plan lifecycle through session-level artifact updates. Routes by mode (plan | verify | replan) and acts as a pure orchestrator: executes phases, parses outputs, and passes context.
+</purpose>

-Unified planning skill combining 4-phase planning workflow, plan quality verification, and interactive replanning. Produces IMPL_PLAN.md, task JSONs, verification reports, and manages plan lifecycle through session-level artifact updates.
+<process>

-## Architecture Overview
+## 1. Architecture Overview

 ```
 ┌──────────────────────────────────────────────────────────────────┐
 │  Workflow Plan Orchestrator (SKILL.md)                            │
 │  → Route by mode: plan | verify | replan                         │
 │  → Pure coordinator: Execute phases, parse outputs, pass context │
-└──────────────────────────────┬───────────────────────────────────┘
+└──────────────────────────────────┬───────────────────────────────┘
                                │
        ┌───────────────────────┼───────────────────────┐
        ↓                       ↓                       ↓
@@ -38,7 +40,7 @@ Unified planning skill combining 4-phase planning workflow, plan quality verific
              └───────────┘─── Review ──→ Display session status inline
 ```

-## Key Design Principles
+## 2. Key Design Principles

 1. **Pure Orchestrator**: SKILL.md routes and coordinates only; execution detail lives in phase files
 2. **Progressive Phase Loading**: Read phase docs ONLY when that phase is about to execute
@@ -47,7 +49,7 @@ Unified planning skill combining 4-phase planning workflow, plan quality verific
 5. **Auto-Continue**: After each phase completes, automatically execute next pending phase
 6. **Accumulated State**: planning-notes.md carries context across phases for N+1 decisions

-## Interactive Preference Collection
+## 3. Interactive Preference Collection

 Before dispatching to phase execution, collect workflow preferences via AskUserQuestion:

@@ -99,7 +101,7 @@ if (autoYes) {

 **workflowPreferences** is passed to phase execution as context variable, referenced as `workflowPreferences.autoYes`, `workflowPreferences.interactive` within phases.

-## Mode Detection
+## 4. Mode Detection

 ```javascript
 const args = $ARGUMENTS
@@ -113,7 +115,7 @@ function detectMode(args) {
 }
 ```

-## Compact Recovery (Phase Persistence)
+## 5. Compact Recovery (Phase Persistence)

 Multi-phase planning (Phase 1-4/5/6) spans long conversations. Uses **双重保险**: TodoWrite 跟踪 active phase 保护其不被压缩，sentinel 作为兜底。

@@ -121,7 +123,7 @@ Multi-phase planning (Phase 1-4/5/6) spans long conversations. Uses **双重保
 > The phase currently marked `in_progress` is the active execution phase — preserve its FULL content.
 > Only compress phases marked `completed` or `pending`.

-## Execution Flow
+## 6. Execution Flow

 ### Plan Mode (default)

@@ -130,23 +132,23 @@ Input Parsing:
   └─ Convert user input to structured format (GOAL/SCOPE/CONTEXT)

 Phase 1: Session Discovery
-   └─ Ref: phases/01-session-discovery.md
+   └─ Ref: Read("phases/01-session-discovery.md")
      └─ Output: sessionId (WFS-xxx), planning-notes.md

 Phase 2: Context Gathering
-   └─ Ref: phases/02-context-gathering.md
+   └─ Ref: Read("phases/02-context-gathering.md")
      ├─ Tasks attached: Analyze structure → Identify integration → Generate package
      └─ Output: contextPath + conflictRisk

 Phase 3: Conflict Resolution (conditional: conflictRisk ≥ medium)
   └─ Decision (conflictRisk check):
-      ├─ conflictRisk ≥ medium → Ref: phases/03-conflict-resolution.md
+      ├─ conflictRisk ≥ medium → Ref: Read("phases/03-conflict-resolution.md")
      │   ├─ Tasks attached: Detect conflicts → Present to user → Apply strategies
      │   └─ Output: Modified brainstorm artifacts
      └─ conflictRisk < medium → Skip to Phase 4

 Phase 4: Task Generation
-   └─ Ref: phases/04-task-generation.md
+   └─ Ref: Read("phases/04-task-generation.md")
      └─ Output: IMPL_PLAN.md, task JSONs, TODO_LIST.md

 Plan Confirmation (User Decision Gate):
@@ -160,7 +162,7 @@ Plan Confirmation (User Decision Gate):

 ```
 Phase 5: Plan Verification
-   └─ Ref: phases/05-plan-verify.md
+   └─ Ref: Read("phases/05-plan-verify.md")
      └─ Output: PLAN_VERIFICATION.md with quality gate recommendation
 ```

@@ -168,7 +170,7 @@ Phase 5: Plan Verification

 ```
 Phase 6: Interactive Replan
-   └─ Ref: phases/06-replan.md
+   └─ Ref: Read("phases/06-replan.md")
      └─ Output: Updated IMPL_PLAN.md, task JSONs, TODO_LIST.md
 ```

@@ -176,19 +178,19 @@ Phase 6: Interactive Replan

 | Phase | Document | Purpose | Mode | Compact |
 |-------|----------|---------|------|---------|
-| 1 | [phases/01-session-discovery.md](phases/01-session-discovery.md) | Create or discover workflow session | plan | TodoWrite 驱动 |
-| 2 | [phases/02-context-gathering.md](phases/02-context-gathering.md) | Gather project context and analyze codebase | plan | TodoWrite 驱动 |
-| 3 | [phases/03-conflict-resolution.md](phases/03-conflict-resolution.md) | Detect and resolve conflicts (conditional) | plan | TodoWrite 驱动 |
-| 4 | [phases/04-task-generation.md](phases/04-task-generation.md) | Generate implementation plan and task JSONs | plan | TodoWrite 驱动 + 🔄 sentinel |
-| 5 | [phases/05-plan-verify.md](phases/05-plan-verify.md) | Read-only verification with quality gate | verify | TodoWrite 驱动 |
-| 6 | [phases/06-replan.md](phases/06-replan.md) | Interactive replanning with boundary clarification | replan | TodoWrite 驱动 |
+| 1 | phases/01-session-discovery.md | Create or discover workflow session | plan | TodoWrite 驱动 |
+| 2 | phases/02-context-gathering.md | Gather project context and analyze codebase | plan | TodoWrite 驱动 |
+| 3 | phases/03-conflict-resolution.md | Detect and resolve conflicts (conditional) | plan | TodoWrite 驱动 |
+| 4 | phases/04-task-generation.md | Generate implementation plan and task JSONs | plan | TodoWrite 驱动 + 🔄 sentinel |
+| 5 | phases/05-plan-verify.md | Read-only verification with quality gate | verify | TodoWrite 驱动 |
+| 6 | phases/06-replan.md | Interactive replanning with boundary clarification | replan | TodoWrite 驱动 |

 **Compact Rules**:
 1. **TodoWrite `in_progress`** → 保留完整内容，禁止压缩
 2. **TodoWrite `completed`** → 可压缩为摘要
 3. **🔄 sentinel fallback** → Phase 4 包含 compact sentinel；若 compact 后仅存 sentinel 而无完整 Step 协议，必须立即 `Read("phases/04-task-generation.md")` 恢复

-## Core Rules
+## 7. Core Rules

 1. **Start Immediately**: First action is mode detection + TodoWrite initialization, second action is phase execution
 2. **No Preliminary Analysis**: Do not read files, analyze structure, or gather context before Phase 1
@@ -199,7 +201,7 @@ Phase 6: Interactive Replan
 7. **Progressive Phase Loading**: Read phase docs ONLY when that phase is about to execute
 8. **DO NOT STOP**: Continuous multi-phase workflow. After executing all attached tasks, immediately collapse them and execute next phase

-## Input Processing
+## 8. Input Processing

 **Convert User Input to Structured Format**:

@@ -228,7 +230,7 @@ Phase 6: Interactive Replan
   - Extract goal, scope, requirements
   - Format into structured description

-## Data Flow
+## 9. Data Flow

 ### Plan Mode

@@ -290,7 +292,7 @@ Phase 6: Mode detection → Clarification → Impact analysis → Backup → App
    ↓ Output: Updated artifacts + change summary
 ```

-## TodoWrite Pattern
+## 10. TodoWrite Pattern

 **Core Concept**: Dynamic task attachment and collapse for real-time visibility into workflow execution.

@@ -349,7 +351,7 @@ Phase 6: Mode detection → Clarification → Impact analysis → Backup → App

 **Note**: See individual Phase descriptions for detailed TodoWrite Update examples.

-## Post-Phase Updates
+## 11. Post-Phase Updates

 After each phase completes, update planning-notes.md:

@@ -360,7 +362,7 @@ After each phase completes, update planning-notes.md:

 See phase files for detailed update code.

-## Error Handling
+## 12. Error Handling

 - **Parsing Failure**: If output parsing fails, retry command once, then report error
 - **Validation Failure**: If validation fails, report which file/data is missing
@@ -368,7 +370,7 @@ See phase files for detailed update code.
 - **Session Not Found** (verify/replan): Report error with available sessions list
 - **Task Not Found** (replan): Report error with available tasks list

-## Coordinator Checklist
+## 13. Coordinator Checklist

 ### Plan Mode
 - **Pre-Phase**: Convert user input to structured format (GOAL/SCOPE/CONTEXT)
@@ -403,7 +405,7 @@ See phase files for detailed update code.
 - Initialize TodoWrite with replan-specific tasks
 - Execute Phase 6 through all sub-phases (clarification → impact → backup → apply → verify)

-## Structure Template Reference
+## 14. Structure Template Reference

 **Minimal Structure**:
 ```
@@ -421,7 +423,7 @@ REQUIREMENTS: [Specific technical requirements]
 CONSTRAINTS: [Limitations or boundaries]
 ```

-## Related Skills
+## 15. Related Skills

 **Prerequisite Skills**:
 - `brainstorm` skill - Optional: Generate role-based analyses before planning
@@ -439,3 +441,26 @@ CONSTRAINTS: [Limitations or boundaries]
 - Display session status inline - Review task breakdown and current progress
 - `Skill(skill="workflow-execute")` - Begin implementation of generated tasks (skill: workflow-execute)
 - `workflow-plan` skill (replan phase) - Modify plan (can also invoke via replan mode)
+
+</process>
+
+<auto_mode>
+When `workflowPreferences.autoYes` is true (triggered by `-y` or `--yes` flag, or user selecting "Auto" mode):
+- Skip all confirmation prompts, use default values
+- Auto-select "Verify Plan Quality" at Plan Confirmation Gate
+- Auto-continue to execution if verification returns PROCEED
+- Skip interactive clarification in replan mode (use safe defaults)
+</auto_mode>
+
+<success_criteria>
+- [ ] Mode correctly detected from skill trigger (plan / verify / replan)
+- [ ] TodoWrite initialized and updated after each phase with attachment/collapse pattern
+- [ ] All phases executed in sequence with proper data passing between phases
+- [ ] Phase documents loaded progressively via Read() only when phase is about to execute
+- [ ] planning-notes.md updated after each phase with accumulated context
+- [ ] Phase 3 conditionally executed based on conflictRisk assessment
+- [ ] Plan Confirmation Gate presented with correct options after Phase 4
+- [ ] All output artifacts generated: IMPL_PLAN.md, task JSONs, TODO_LIST.md
+- [ ] Compact recovery works: in_progress phases preserved, completed phases compressible
+- [ ] Error handling covers parsing failures, validation failures, and missing sessions/tasks
+</success_criteria>
--- a/.claude/skills/workflow-tdd-plan/SKILL.md
+++ b/.claude/skills/workflow-tdd-plan/SKILL.md
@@ -4,18 +4,20 @@ description: Unified TDD workflow skill combining 6-phase TDD planning with Red-
 allowed-tools: Skill, Agent, AskUserQuestion, TaskCreate, TaskUpdate, TaskList, Read, Write, Edit, Bash, Glob, Grep
 ---

-# Workflow TDD
+<purpose>
+Unified TDD workflow skill combining TDD planning (Red-Green-Refactor task chain generation with test-first development structure) and TDD verification (compliance validation with quality gate reporting). Produces IMPL_PLAN.md, task JSONs with internal TDD cycles, and TDD_COMPLIANCE_REPORT.md. Triggers on "workflow-tdd-plan" (plan mode) or "workflow-tdd-verify" (verify mode).
+</purpose>

-Unified TDD workflow skill combining TDD planning (Red-Green-Refactor task chain generation with test-first development structure) and TDD verification (compliance validation with quality gate reporting). Produces IMPL_PLAN.md, task JSONs with internal TDD cycles, and TDD_COMPLIANCE_REPORT.md.
+<process>

-## Architecture Overview
+## 1. Architecture Overview

 ```
 ┌──────────────────────────────────────────────────────────────────┐
 │  Workflow TDD Orchestrator (SKILL.md)                            │
 │  → Route by mode: plan | verify                                  │
 │  → Pure coordinator: Execute phases, parse outputs, pass context │
-└──────────────────────────────┬───────────────────────────────────┘
+└──────────────────────────────────┬───────────────────────────────┘
                                │
        ┌───────────────────────┴───────────────────────┐
        ↓                                               ↓
@@ -38,7 +40,7 @@ Unified TDD workflow skill combining TDD planning (Red-Green-Refactor task chain
                       └───────────┘─── Review ──→ Display session status inline
 ```

-## Key Design Principles
+## 2. Key Design Principles

 1. **Pure Orchestrator**: SKILL.md routes and coordinates only; execution detail lives in phase files
 2. **Progressive Phase Loading**: Read phase docs ONLY when that phase is about to execute
@@ -47,7 +49,7 @@ Unified TDD workflow skill combining TDD planning (Red-Green-Refactor task chain
 5. **Auto-Continue**: After each phase completes, automatically execute next pending phase
 6. **TDD Iron Law**: NO PRODUCTION CODE WITHOUT A FAILING TEST FIRST - enforced in task structure

-## Interactive Preference Collection
+## 3. Interactive Preference Collection

 Before dispatching to phase execution, collect workflow preferences via AskUserQuestion:

@@ -81,7 +83,7 @@ if (autoYes) {

 **workflowPreferences** is passed to phase execution as context variable, referenced as `workflowPreferences.autoYes` within phases.

-## Mode Detection
+## 4. Mode Detection

 ```javascript
 const args = $ARGUMENTS
@@ -94,7 +96,7 @@ function detectMode(args) {
 }
 ```

-## Compact Recovery (Phase Persistence)
+## 5. Compact Recovery (Phase Persistence)

 Multi-phase TDD planning (Phase 1-6/7) spans long conversations. Uses **双重保险**: TodoWrite 跟踪 active phase 保护其不被压缩，sentinel 作为兜底。

@@ -102,42 +104,40 @@ Multi-phase TDD planning (Phase 1-6/7) spans long conversations. Uses **双重
 > The phase currently marked `in_progress` is the active execution phase — preserve its FULL content.
 > Only compress phases marked `completed` or `pending`.

-## Execution Flow
-
-### Plan Mode (default)
+## 6. Execution Flow — Plan Mode (default)

 ```
 Input Parsing:
   └─ Convert user input to TDD structured format (GOAL/SCOPE/CONTEXT/TEST_FOCUS)

 Phase 1: Session Discovery
-   └─ Ref: phases/01-session-discovery.md
+   └─ Read("phases/01-session-discovery.md")
      └─ Output: sessionId (WFS-xxx)

 Phase 2: Context Gathering
-   └─ Ref: phases/02-context-gathering.md
+   └─ Read("phases/02-context-gathering.md")
      ├─ Tasks attached: Analyze structure → Identify integration → Generate package
      └─ Output: contextPath + conflictRisk

 Phase 3: Test Coverage Analysis
-   └─ Ref: phases/03-test-coverage-analysis.md
+   └─ Read("phases/03-test-coverage-analysis.md")
      ├─ Tasks attached: Detect framework → Analyze coverage → Identify gaps
      └─ Output: testContextPath

 Phase 4: Conflict Resolution (conditional: conflictRisk ≥ medium)
   └─ Decision (conflictRisk check):
-      ├─ conflictRisk ≥ medium → Ref: phases/04-conflict-resolution.md
+      ├─ conflictRisk ≥ medium → Read("phases/04-conflict-resolution.md")
      │   ├─ Tasks attached: Detect conflicts → Log analysis → Apply strategies
      │   └─ Output: conflict-resolution.json
      └─ conflictRisk < medium → Skip to Phase 5

 Phase 5: TDD Task Generation
-   └─ Ref: phases/05-tdd-task-generation.md
+   └─ Read("phases/05-tdd-task-generation.md")
      ├─ Tasks attached: Discovery → Planning → Output
      └─ Output: IMPL_PLAN.md, IMPL-*.json, TODO_LIST.md

 Phase 6: TDD Structure Validation
-   └─ Ref: phases/06-tdd-structure-validation.md
+   └─ Read("phases/06-tdd-structure-validation.md")
      └─ Output: Validation report + Plan Confirmation Gate

 Plan Confirmation (User Decision Gate):
@@ -147,32 +147,34 @@ Plan Confirmation (User Decision Gate):
      └─ "Review Status Only" → Display session status inline
 ```

-### Verify Mode
+## 7. Execution Flow — Verify Mode

 ```
 Phase 7: TDD Verification
-   └─ Ref: phases/07-tdd-verify.md
+   └─ Read("phases/07-tdd-verify.md")
      └─ Output: TDD_COMPLIANCE_REPORT.md with quality gate recommendation
 ```

-**Phase Reference Documents** (read on-demand when phase executes):
+## 8. Phase Reference Documents
+
+Read on-demand when phase executes using `Read("phases/...")`:

 | Phase | Document | Purpose | Mode | Compact |
 |-------|----------|---------|------|---------|
-| 1 | [phases/01-session-discovery.md](phases/01-session-discovery.md) | Create or discover TDD workflow session | plan | TodoWrite 驱动 |
-| 2 | [phases/02-context-gathering.md](phases/02-context-gathering.md) | Gather project context and analyze codebase | plan | TodoWrite 驱动 |
-| 3 | [phases/03-test-coverage-analysis.md](phases/03-test-coverage-analysis.md) | Analyze test coverage and framework detection | plan | TodoWrite 驱动 |
-| 4 | [phases/04-conflict-resolution.md](phases/04-conflict-resolution.md) | Detect and resolve conflicts (conditional) | plan | TodoWrite 驱动 |
-| 5 | [phases/05-tdd-task-generation.md](phases/05-tdd-task-generation.md) | Generate TDD tasks with Red-Green-Refactor cycles | plan | TodoWrite 驱动 + 🔄 sentinel |
-| 6 | [phases/06-tdd-structure-validation.md](phases/06-tdd-structure-validation.md) | Validate TDD structure and present confirmation gate | plan | TodoWrite 驱动 + 🔄 sentinel |
-| 7 | [phases/07-tdd-verify.md](phases/07-tdd-verify.md) | Full TDD compliance verification with quality gate | verify | TodoWrite 驱动 |
+| 1 | phases/01-session-discovery.md | Create or discover TDD workflow session | plan | TodoWrite 驱动 |
+| 2 | phases/02-context-gathering.md | Gather project context and analyze codebase | plan | TodoWrite 驱动 |
+| 3 | phases/03-test-coverage-analysis.md | Analyze test coverage and framework detection | plan | TodoWrite 驱动 |
+| 4 | phases/04-conflict-resolution.md | Detect and resolve conflicts (conditional) | plan | TodoWrite 驱动 |
+| 5 | phases/05-tdd-task-generation.md | Generate TDD tasks with Red-Green-Refactor cycles | plan | TodoWrite 驱动 + sentinel |
+| 6 | phases/06-tdd-structure-validation.md | Validate TDD structure and present confirmation gate | plan | TodoWrite 驱动 + sentinel |
+| 7 | phases/07-tdd-verify.md | Full TDD compliance verification with quality gate | verify | TodoWrite 驱动 |

 **Compact Rules**:
 1. **TodoWrite `in_progress`** → 保留完整内容，禁止压缩
 2. **TodoWrite `completed`** → 可压缩为摘要
-3. **🔄 sentinel fallback** → Phase 5/6 包含 compact sentinel；若 compact 后仅存 sentinel 而无完整 Step 协议，必须立即 `Read()` 恢复对应 phase 文件
+3. **sentinel fallback** → Phase 5/6 包含 compact sentinel；若 compact 后仅存 sentinel 而无完整 Step 协议，必须立即 `Read()` 恢复对应 phase 文件

-## Core Rules
+## 9. Core Rules

 1. **Start Immediately**: First action is mode detection + TaskCreate initialization, second action is phase execution
 2. **No Preliminary Analysis**: Do not read files, analyze structure, or gather context before Phase 1
@@ -184,7 +186,7 @@ Phase 7: TDD Verification
 8. **DO NOT STOP**: Continuous multi-phase workflow. After executing all attached tasks, immediately collapse them and execute next phase
 9. **TDD Context**: All descriptions include "TDD:" prefix

-## TDD Compliance Requirements
+## 10. TDD Compliance Requirements

 ### The Iron Law

@@ -222,7 +224,7 @@ NO PRODUCTION CODE WITHOUT A FAILING TEST FIRST
 - Test-first forces edge case discovery before implementation
 - Tests-after verify what was built, not what's required

-## Input Processing
+## 11. Input Processing

 **Convert User Input to TDD Structured Format**:

@@ -252,9 +254,7 @@ NO PRODUCTION CODE WITHOUT A FAILING TEST FIRST

 3. **File/Issue** → Read and structure with TDD

-## Data Flow
-
-### Plan Mode
+## 12. Data Flow — Plan Mode

 ```
 User Input (task description)
@@ -297,7 +297,7 @@ Plan Confirmation (User Decision Gate):
    └─ "Review Status Only" → Display session status inline
 ```

-### Verify Mode
+## 13. Data Flow — Verify Mode

 ```
 Input: --session sessionId (or auto-detect)
@@ -311,7 +311,7 @@ Phase 7: Session discovery → Chain validation → Coverage analysis → Report
 - Existing context and analysis
 - Session-specific configuration

-## TodoWrite Pattern
+## 14. TodoWrite Pattern

 **Core Concept**: Dynamic task attachment and collapse for real-time visibility into TDD workflow execution.

@@ -394,7 +394,7 @@ Phase 7: Session discovery → Chain validation → Coverage analysis → Report

 **Note**: See individual Phase descriptions for detailed TodoWrite Update examples.

-## Post-Phase Updates
+## 15. Post-Phase Updates

 ### Memory State Check

@@ -409,7 +409,7 @@ After heavy phases (Phase 2-3), evaluate context window usage:

 Similar to workflow-plan, a `planning-notes.md` can accumulate context across phases if needed. See Phase 1 for initialization.

-## Error Handling
+## 16. Error Handling

 - **Parsing Failure**: If output parsing fails, retry command once, then report error
 - **Validation Failure**: Report which file/data is missing or invalid
@@ -447,9 +447,8 @@ Similar to workflow-plan, a `planning-notes.md` can accumulate context across ph
 2. Summary displayed in Phase 6 output
 3. User decides whether to address before `workflow-execute` skill

-## Coordinator Checklist
+## 17. Coordinator Checklist — Plan Mode

-### Plan Mode
 - **Pre-Phase**: Convert user input to TDD structured format (TDD/GOAL/SCOPE/CONTEXT/TEST_FOCUS)
 - Initialize TaskCreate before any command (Phase 4 added dynamically after Phase 2)
 - Execute Phase 1 immediately with structured description
@@ -466,20 +465,21 @@ Similar to workflow-plan, a `planning-notes.md` can accumulate context across ph
 - Verify all Phase 5 outputs (IMPL_PLAN.md, IMPL-*.json, TODO_LIST.md)
 - Execute Phase 6 (internal TDD structure validation)
 - **Plan Confirmation Gate**: Present user with choice (Verify → Phase 7 / Execute / Review Status)
- **If user selects Verify**: Read phases/07-tdd-verify.md, execute Phase 7 in-process
+- **If user selects Verify**: Read("phases/07-tdd-verify.md"), execute Phase 7 in-process
 - **If user selects Execute**: Skill(skill="workflow-execute")
 - **If user selects Review**: Display session status inline
 - **Auto mode (workflowPreferences.autoYes)**: Auto-select "Verify TDD Compliance", then auto-continue to execute if APPROVED
 - Update TaskCreate/TaskUpdate after each phase
 - After each phase, automatically continue to next phase based on TaskList status

-### Verify Mode
+## 18. Coordinator Checklist — Verify Mode
+
 - Detect/validate session (from --session flag or auto-detect)
 - Initialize TaskCreate with verification tasks
 - Execute Phase 7 through all sub-phases (session validation → chain validation → coverage analysis → report generation)
 - Present quality gate result and next step options

-## Related Skills
+## 19. Related Skills

 **Prerequisite Skills**:
 - None - TDD planning is self-contained (can optionally run brainstorm commands before)
@@ -500,3 +500,28 @@ Similar to workflow-plan, a `planning-notes.md` can accumulate context across ph
 - `workflow-plan` skill (plan-verify phase) - Verify plan quality and dependencies
 - Display session status inline - Review TDD task breakdown
 - `Skill(skill="workflow-execute")` - Begin TDD implementation
+
+</process>
+
+<auto_mode>
+When `workflowPreferences.autoYes` is true (triggered by `-y`/`--yes` flag):
+- Skip all interactive confirmation prompts
+- Use default values for all preference questions
+- At Plan Confirmation Gate: Auto-select "Verify TDD Compliance"
+- After verification: Auto-continue to execute if quality gate returns APPROVED
+- All phases execute continuously without user intervention
+</auto_mode>
+
+<success_criteria>
+- [ ] Mode correctly detected from skill trigger name (plan vs verify)
+- [ ] All 6 plan phases execute sequentially with proper data flow between them
+- [ ] Phase files loaded progressively via Read() only when phase is about to execute
+- [ ] TaskCreate/TaskUpdate tracks all phases with attachment/collapse pattern
+- [ ] TDD Iron Law enforced: every task has Red-Green-Refactor structure
+- [ ] Phase 4 (Conflict Resolution) conditionally executes based on conflictRisk level
+- [ ] Plan Confirmation Gate presents three choices after Phase 6
+- [ ] Verify mode (Phase 7) produces TDD_COMPLIANCE_REPORT.md with quality gate
+- [ ] All outputs generated: IMPL_PLAN.md, IMPL-*.json, TODO_LIST.md
+- [ ] Compact recovery preserves active phase content via TodoWrite status
+- [ ] Error handling retries once on parsing failure, reports on persistent errors
+</success_criteria>
--- a/.claude/skills/workflow-test-fix/SKILL.md
+++ b/.claude/skills/workflow-test-fix/SKILL.md
@@ -4,11 +4,13 @@ description: Unified test-fix pipeline combining test generation (session, conte
 allowed-tools: Skill, Agent, AskUserQuestion, TaskCreate, TaskUpdate, TaskList, Read, Write, Edit, Bash, Glob, Grep
 ---

-# Workflow Test Fix
+<purpose>
+Unified test-fix orchestrator that combines **test planning generation** (Phase 1-4) with **iterative test-cycle execution** (Phase 5) into a single end-to-end pipeline. Creates test sessions with progressive L0-L3 test layers, generates test tasks, then executes them with adaptive fix cycles until pass rate >= 95% or max iterations reached. Triggered via skill name routing for full pipeline or execute-only modes.
+</purpose>

-Unified test-fix orchestrator that combines **test planning generation** (Phase 1-4) with **iterative test-cycle execution** (Phase 5) into a single end-to-end pipeline. Creates test sessions with progressive L0-L3 test layers, generates test tasks, then executes them with adaptive fix cycles until pass rate >= 95% or max iterations reached.
+<process>

-## Architecture Overview
+## 1. Architecture Overview

 ```
 ┌───────────────────────────────────────────────────────────────────────────┐
@@ -45,7 +47,7 @@ Task Pipeline (generated in Phase 4, executed in Phase 5):
 └──────────────┘   └─────────────────┘   └─────────────────┘   └──────────────┘
 ```

-## Key Design Principles
+## 2. Key Design Principles

 1. **Unified Pipeline**: Generation and execution are one continuous workflow - no manual handoff
 2. **Pure Orchestrator**: SKILL.md coordinates only - delegates all execution detail to phase files
@@ -56,14 +58,14 @@ Task Pipeline (generated in Phase 4, executed in Phase 5):
 7. **Quality Gate**: Pass rate >= 95% (criticality-aware) terminates the fix loop
 8. **Phase File Hygiene**: Phase files reference `workflowPreferences.*` for preferences, no CLI flag parsing

-## Usage
+## 3. Usage

 Full pipeline and execute-only modes are triggered by skill name routing (see Mode Detection). Workflow preferences (auto mode) are collected interactively via AskUserQuestion before dispatching to phases.

 **Full pipeline** (workflow-test-fix): Task description or session ID as arguments → interactive preference collection → generate + execute pipeline
 **Execute only** (workflow-test-fix): Auto-discovers active session → interactive preference collection → execution loop

-## Interactive Preference Collection
+## 4. Interactive Preference Collection

 Before dispatching to phase execution, collect workflow preferences via AskUserQuestion:

@@ -97,7 +99,7 @@ if (autoYes) {

 **workflowPreferences** is passed to phase execution as context variable, referenced as `workflowPreferences.autoYes` within phases.

-## Compact Recovery (Phase Persistence)
+## 5. Compact Recovery (Phase Persistence)

 Multi-phase test-fix pipeline (Phase 1-5) spans long conversations, especially Phase 5 fix loops. Uses **双重保险**: TodoWrite 跟踪 active phase 保护其不被压缩，sentinel 作为兜底。

@@ -105,7 +107,7 @@ Multi-phase test-fix pipeline (Phase 1-5) spans long conversations, especially P
 > The phase currently marked `in_progress` is the active execution phase — preserve its FULL content.
 > Only compress phases marked `completed` or `pending`.

-## Execution Flow
+## 6. Execution Flow

 ```
 Entry Point Detection:
@@ -113,23 +115,23 @@ Entry Point Detection:
   └─ /workflow-test-fix → Execution Only (Phase 5)

 Phase 1: Session Start (session-start)
-   └─ Ref: phases/01-session-start.md
+   └─ Read("phases/01-session-start.md")
      ├─ Step 1.0: Detect input mode (session | prompt)
      ├─ Step 1.1: Create test session → testSessionId
      └─ Output: testSessionId, MODE

 Phase 2: Test Context Gather (test-context-gather)
-   └─ Ref: phases/02-test-context-gather.md
+   └─ Read("phases/02-test-context-gather.md")
      ├─ Step 1.2: Gather test context → contextPath
      └─ Output: contextPath

 Phase 3: Test Concept Enhanced (test-concept-enhanced)
-   └─ Ref: phases/03-test-concept-enhanced.md
+   └─ Read("phases/03-test-concept-enhanced.md")
      ├─ Step 1.3: Test analysis (Gemini) → TEST_ANALYSIS_RESULTS.md
      └─ Output: TEST_ANALYSIS_RESULTS.md

 Phase 4: Test Task Generate (test-task-generate)
-   └─ Ref: phases/04-test-task-generate.md
+   └─ Read("phases/04-test-task-generate.md")
      ├─ Step 1.4: Generate test tasks → IMPL_PLAN.md, IMPL-*.json, TODO_LIST.md
      └─ Output: testSessionId, 4+ task JSONs

@@ -137,7 +139,7 @@ Summary Output (inline after Phase 4):
   └─ Display summary, auto-continue to Phase 5

 Phase 5: Test Cycle Execution (test-cycle-execute)
-   └─ Ref: phases/05-test-cycle-execute.md
+   └─ Read("phases/05-test-cycle-execute.md")
      ├─ Step 2.1: Discovery (load session, tasks, iteration state)
      ├─ Step 2.2: Execute initial tasks (IMPL-001 → 001.3 → 001.5 → 002)
      ├─ Step 2.3: Fix loop (if pass_rate < 95%)
@@ -153,18 +155,18 @@ Phase 5: Test Cycle Execution (test-cycle-execute)

 | Phase | Document | Purpose | Compact |
 |-------|----------|---------|---------|
-| 1 | [phases/01-session-start.md](phases/01-session-start.md) | Detect input mode, create test session | TodoWrite 驱动 |
-| 2 | [phases/02-test-context-gather.md](phases/02-test-context-gather.md) | Gather test context (coverage/codebase) | TodoWrite 驱动 |
-| 3 | [phases/03-test-concept-enhanced.md](phases/03-test-concept-enhanced.md) | Gemini analysis, L0-L3 test requirements | TodoWrite 驱动 |
-| 4 | [phases/04-test-task-generate.md](phases/04-test-task-generate.md) | Generate task JSONs and IMPL_PLAN.md | TodoWrite 驱动 |
-| 5 | [phases/05-test-cycle-execute.md](phases/05-test-cycle-execute.md) | Execute tasks, iterative fix cycles, completion | TodoWrite 驱动 + 🔄 sentinel |
+| 1 | phases/01-session-start.md | Detect input mode, create test session | TodoWrite 驱动 |
+| 2 | phases/02-test-context-gather.md | Gather test context (coverage/codebase) | TodoWrite 驱动 |
+| 3 | phases/03-test-concept-enhanced.md | Gemini analysis, L0-L3 test requirements | TodoWrite 驱动 |
+| 4 | phases/04-test-task-generate.md | Generate task JSONs and IMPL_PLAN.md | TodoWrite 驱动 |
+| 5 | phases/05-test-cycle-execute.md | Execute tasks, iterative fix cycles, completion | TodoWrite 驱动 + 🔄 sentinel |

 **Compact Rules**:
 1. **TodoWrite `in_progress`** → 保留完整内容，禁止压缩
 2. **TodoWrite `completed`** → 可压缩为摘要
 3. **🔄 sentinel fallback** → Phase 5 包含 compact sentinel；若 compact 后仅存 sentinel 而无完整 Step 协议，必须立即 `Read("phases/05-test-cycle-execute.md")` 恢复

-## Core Rules
+## 7. Core Rules

 1. **Start Immediately**: First action is TaskCreate initialization, second action is Phase 1 (or Phase 5 for execute-only entry)
 2. **No Preliminary Analysis**: Do not read files or gather context before starting the phase
@@ -176,7 +178,7 @@ Phase 5: Test Cycle Execution (test-cycle-execute)
 8. **Progressive Loading**: Read phase doc ONLY when that phase is about to execute
 9. **Entry Point Routing**: `workflow-test-fix` skill → Phase 1-5; `workflow-test-fix` skill → Phase 5 only

-## Input Processing
+## 8. Input Processing

 ### test-fix-gen Entry (Full Pipeline)
 ```
@@ -194,7 +196,7 @@ Arguments → Parse flags:
  └─ (no args)                  → auto-discover active test session
 ```

-## Data Flow
+## 9. Data Flow

 ```
 User Input (session ID | description | file path)
@@ -223,7 +225,7 @@ Phase 5: Test Cycle Execution ────────────────
    ↓ 2.4: Completion → summary → session archive
 ```

-## Summary Output (after Phase 4)
+## 10. Summary Output (after Phase 4)

 After Phase 4 completes, display the following summary before auto-continuing to Phase 5:

@@ -255,7 +257,7 @@ Review artifacts:
 **CRITICAL - Next Step**: Auto-continue to Phase 5: Test Cycle Execution.
 Pass `testSessionId` to Phase 5 for test execution pipeline. Do NOT wait for user confirmation — the unified pipeline continues automatically.

-## Test Strategy Overview
+## 11. Test Strategy Overview

 Progressive Test Layers (L0-L3):

@@ -273,7 +275,7 @@ Progressive Test Layers (L0-L3):
 - Pass Rate Gate: >= 95% (criticality-aware) or 100%
 - Max Fix Iterations: 10 (default, adjustable)

-## Strategy Engine (Phase 5)
+## 12. Strategy Engine (Phase 5)

 | Strategy | Trigger | Behavior |
 |----------|---------|----------|
@@ -283,7 +285,7 @@ Progressive Test Layers (L0-L3):

 Selection logic and CLI fallback chain (Gemini → Qwen → Codex) are detailed in Phase 5.

-## Agent Roles
+## 13. Agent Roles

 | Agent | Used In | Responsibility |
 |-------|---------|---------------|
@@ -292,7 +294,7 @@ Selection logic and CLI fallback chain (Gemini → Qwen → Codex) are detailed
 | **@test-fix-agent** | Phase 5 | Test execution, code fixes, criticality assignment |
 | **@cli-planning-agent** | Phase 5 (fix loop) | CLI analysis, root cause extraction, fix task generation |

-## TodoWrite Pattern
+## 14. TodoWrite Pattern

 **Core Concept**: Dynamic task tracking with attachment/collapse for real-time visibility.

@@ -344,7 +346,7 @@ Selection logic and CLI fallback chain (Gemini → Qwen → Codex) are detailed
 ]
 ```

-## Session File Structure
+## 15. Session File Structure

 ```
 .workflow/active/WFS-test-{session}/
@@ -370,7 +372,7 @@ Selection logic and CLI fallback chain (Gemini → Qwen → Codex) are detailed
    └── iteration-summaries/
 ```

-## Error Handling
+## 16. Error Handling

 ### Phase 1-4 (Generation)

@@ -393,13 +395,13 @@ Selection logic and CLI fallback chain (Gemini → Qwen → Codex) are detailed
 | Regression detected | Rollback last fix, switch to surgical strategy |
 | Stuck tests detected | Continue with alternative strategy, document |

-## Commit Strategy (Phase 5)
+## 17. Commit Strategy (Phase 5)

 Automatic commits at key checkpoints:
 1. **After successful iteration** (pass rate increased): `test-cycle: iteration N - strategy (pass: old% → new%)`
 2. **Before rollback** (regression detected): `test-cycle: rollback iteration N - regression detected`

-## Completion Conditions
+## 18. Completion Conditions

 | Condition | Pass Rate | Action |
 |-----------|-----------|--------|
@@ -407,36 +409,36 @@ Automatic commits at key checkpoints:
 | **Partial Success** | >= 95%, all failures low criticality | Auto-approve with review note |
 | **Failure** | < 95% after max iterations | Failure report, mark blocked |

-## Post-Completion Expansion
+## 19. Post-Completion Expansion

 **Auto-sync**: Execute `/workflow:session:sync -y "{summary}"` to update specs/*.md + project-tech.

 After completion, ask user if they want to expand into issues (test/enhance/refactor/doc). Selected items call `/issue:new "{summary} - {dimension}"`.

-## Coordinator Checklist
+## 20. Coordinator Checklist

 ### Phase 1 (session-start)
 - [ ] Detect input type (session ID / description / file path)
 - [ ] Initialize TaskCreate before any execution
- [ ] Read Phase 1 doc, execute Steps 1.0 + 1.1
+- [ ] Read("phases/01-session-start.md"), execute Steps 1.0 + 1.1
 - [ ] Parse testSessionId from step output, store in memory

 ### Phase 2 (test-context-gather)
- [ ] Read Phase 2 doc, execute Step 1.2
+- [ ] Read("phases/02-test-context-gather.md"), execute Step 1.2
 - [ ] Parse contextPath from step output, store in memory

 ### Phase 3 (test-concept-enhanced)
- [ ] Read Phase 3 doc, execute Step 1.3
+- [ ] Read("phases/03-test-concept-enhanced.md"), execute Step 1.3
 - [ ] Verify TEST_ANALYSIS_RESULTS.md created

 ### Phase 4 (test-task-generate)
- [ ] Read Phase 4 doc, execute Step 1.4
+- [ ] Read("phases/04-test-task-generate.md"), execute Step 1.4
 - [ ] Verify all Phase 1-4 outputs (4 task JSONs, IMPL_PLAN.md, TODO_LIST.md)
 - [ ] Display Summary output (inline)
 - [ ] Collapse Phase 1-4 tasks, auto-continue to Phase 5

 ### Phase 5 (test-cycle-execute)
- [ ] Read Phase 5 doc
+- [ ] Read("phases/05-test-cycle-execute.md")
 - [ ] Load session, tasks, iteration state
 - [ ] Execute initial tasks sequentially
 - [ ] Calculate pass rate from test-results.json
@@ -446,7 +448,7 @@ After completion, ask user if they want to expand into issues (test/enhance/refa
 - [ ] Generate completion summary
 - [ ] Offer post-completion expansion

-## Related Skills
+## 21. Related Skills

 **Prerequisite Skills**:
 - `workflow-plan` skill or `workflow-execute` skill - Complete implementation (Session Mode source)
@@ -456,3 +458,25 @@ After completion, ask user if they want to expand into issues (test/enhance/refa
 - Display session status inline - Review workflow state
 - `review-cycle` skill - Post-implementation review
 - `/issue:new` - Create follow-up issues
+
+</process>
+
+<auto_mode>
+When `-y` or `--yes` is detected in $ARGUMENTS or propagated via ccw:
+- Skip all AskUserQuestion confirmations
+- Use default values for all workflow preferences (`workflowPreferences = { autoYes: true }`)
+- Auto-continue through all phases without user interaction
+- Phase 1→2→3→4→Summary→5 executes as a fully automatic pipeline
+</auto_mode>
+
+<success_criteria>
+- [ ] Input type correctly detected (session ID / description / file path)
+- [ ] All 5 phases execute in sequence (full pipeline) or Phase 5 only (execute-only)
+- [ ] Phase documents loaded progressively via Read() only when phase executes
+- [ ] TaskCreate/TaskUpdate tracking maintained throughout with attachment/collapse pattern
+- [ ] All phase outputs parsed and passed to subsequent phases (testSessionId, contextPath, etc.)
+- [ ] Summary displayed after Phase 4 with all task and threshold details
+- [ ] Phase 5 fix loop iterates with adaptive strategy until pass rate >= 95% or max iterations
+- [ ] Completion summary generated with final pass rate and session archived
+- [ ] Post-completion expansion offered to user
+</success_criteria>