Refactor implementation approach structure across task generation files

- Updated implementation_approach in task-generate-tdd.md to use an array format with detailed steps for minimal code implementation and iterative testing. - Enhanced task-generate.md to support CLI execution mode with Codex, including examples for task execution and session management. - Modified test-task-generate.md to incorporate CLI execution mode, allowing Codex to autonomously generate tests and execute iterative test-fix cycles. - Revised task-core.md and workflow-architecture.md to adopt a step-based implementation approach, improving clarity and dependency management in task execution. - Improved documentation in workflow-architecture.md to clarify flow control structure and variable referencing.
2026-02-12 02:37:45 +08:00 · 2025-10-12 14:09:47 +08:00
parent c5aac313ff
commit 58153ecb83
15 changed files with 1479 additions and 920 deletions
--- a/.claude/commands/workflow/execute.md
+++ b/.claude/commands/workflow/execute.md
@@ -425,10 +425,28 @@ Task(subagent_type="{meta.agent}",
        "on_error": "skip_optional|fail|retry_once"
      }
    ],
-    "implementation_approach": {
-      "task_description": "Implement following consolidated synthesis specification...",
-      "modification_points": ["Apply synthesis specification requirements..."]
-    },
+    "implementation_approach": [
+      {
+        "step": 1,
+        "title": "Implement task following synthesis specification",
+        "description": "Implement '[title]' following synthesis specification. PRIORITY: Use synthesis-specification.md 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 synthesis-specification.md",
+          "Follow technical guidelines from synthesis",
+          "Consult artifacts for implementation details when needed",
+          "Integrate with existing patterns"
+        ],
+        "logic_flow": [
+          "Load synthesis specification",
+          "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"]
  }
 }
--- a/.claude/commands/workflow/plan.md
+++ b/.claude/commands/workflow/plan.md
@@ -1,7 +1,7 @@
 ---
 name: plan
 description: Orchestrate 4-phase planning workflow by executing commands and passing context between phases
-argument-hint: "[--agent] \"text description\"|file.md"
+argument-hint: "[--agent] [--cli-execute] \"text description\"|file.md"
 allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 ---

@@ -30,6 +30,7 @@ This workflow runs **fully autonomously** once triggered. Each phase completes,
 **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

 ## Core Rules

@@ -118,9 +119,23 @@ CONTEXT: Existing user database schema, REST API endpoints
 - **IMPL_PLAN.md defines "HOW"**: Executable task breakdown, dependencies, implementation sequence
 - Task generation translates high-level specifications into concrete, actionable work items

-**Command**:
+**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
+
+**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
+```

 **Input**: `sessionId` from Phase 1

@@ -254,7 +269,10 @@ Return summary to user
 ✅ Parse context path from Phase 2 output, store in memory
 ✅ Pass session ID and context path to Phase 3 command
 ✅ Verify ANALYSIS_RESULTS.md after Phase 3
-✅ Select correct Phase 4 command based on --agent flag
+✅ **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
 ✅ Pass session ID to Phase 4 command
 ✅ Verify all Phase 4 outputs
 ✅ Update TodoWrite after each phase
--- a/.claude/commands/workflow/test-gen.md
+++ b/.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] source-session-id"
+argument-hint: "[--use-codex] [--cli-execute] source-session-id"
 allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 ---

@@ -123,11 +123,12 @@ allowed-tools: SlashCommand(*), TodoWrite(*), Read(*), Bash(*)
 ---

 ### Phase 4: Generate Test Tasks
-**Command**: `SlashCommand(command="/workflow:tools:test-task-generate [--use-codex] --session [testSessionId]")`
+**Command**: `SlashCommand(command="/workflow:tools:test-task-generate [--use-codex] [--cli-execute] --session [testSessionId]")`

 **Input**:
 - `testSessionId` from Phase 1
- `--use-codex` flag (if present in original command)
+- `--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

 **Expected Behavior**:
 - Parse TEST_ANALYSIS_RESULTS.md from Phase 3
--- a/.claude/commands/workflow/tools/docs.md
+++ b/.claude/commands/workflow/tools/docs.md
--- a/.claude/commands/workflow/tools/task-generate-agent.md
+++ b/.claude/commands/workflow/tools/task-generate-agent.md
@@ -215,17 +215,29 @@ Task(
        "on_error": "fail"
      }
    ],
-    "implementation_approach": {
-      "task_description": "Implement '[title]' following synthesis specification. PRIORITY: Use synthesis-specification.md 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 requirements from synthesis"],
-      "logic_flow": [
-        "Load synthesis specification",
-        "Analyze existing patterns",
-        "Implement following specification",
-        "Consult artifacts for technical details when needed",
-        "Validate against acceptance criteria"
-      ]
-    },
+    "implementation_approach": [
+      {
+        "step": 1,
+        "title": "Implement task following synthesis specification",
+        "description": "Implement '[title]' following synthesis specification. PRIORITY: Use synthesis-specification.md 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 synthesis-specification.md",
+          "Follow technical guidelines from synthesis",
+          "Consult artifacts for implementation details when needed",
+          "Integrate with existing patterns"
+        ],
+        "logic_flow": [
+          "Load synthesis specification and relevant role artifacts",
+          "Execute MCP code-index discovery for relevant files",
+          "Analyze existing patterns and identify modification targets",
+          "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"]
  }
 }
--- a/.claude/commands/workflow/tools/task-generate-tdd.md
+++ b/.claude/commands/workflow/tools/task-generate-tdd.md
@@ -174,51 +174,53 @@ For each feature, generate 3 tasks with ID format:
        "on_error": "skip_optional"
      }
    ],
-    "implementation_approach": {
-      "task_description": "Write minimal code to pass tests, then enter iterative fix cycle if they still fail",
-      "initial_implementation": [
-        "Write minimal code based on test requirements",
-        "Execute test suite: bash(npm test -- tests/auth/login.test.ts)",
-        "If tests pass → Complete task",
-        "If tests fail → Capture failure logs and proceed to test-fix cycle"
-      ],
-      "test_fix_cycle": {
-        "max_iterations": 3,
-        "cycle_pattern": "gemini_diagnose → manual_fix (or codex if meta.use_codex=true) → retest",
-        "tools": {
-          "diagnosis": "gemini-wrapper (MODE: analysis, uses bug-fix template)",
-          "fix_application": "manual (default) or codex if meta.use_codex=true",
-          "verification": "bash(npm test -- tests/auth/login.test.ts)"
-        },
-        "exit_conditions": {
-          "success": "all_tests_pass",
-          "failure": "max_iterations_reached"
-        },
-        "steps": [
-          "ITERATION LOOP (max 3):",
-          "  1. Gemini Diagnosis:",
-          "     bash(cd .workflow/WFS-xxx/.process && ~/.claude/scripts/gemini-wrapper --all-files -p \"",
-          "     PURPOSE: Diagnose TDD Green phase test failure iteration [N]",
-          "     TASK: Systematic bug analysis and fix recommendations",
-          "     MODE: analysis",
-          "     CONTEXT: @{CLAUDE.md,**/*CLAUDE.md}",
-          "              Test output: [test_failures]",
-          "              Test requirements: [test_requirements]",
-          "              Implementation: [focus_paths]",
-          "     EXPECTED: Root cause analysis, code path tracing, targeted fixes",
-          "     RULES: $(cat ~/.claude/prompt-templates/bug-fix.md) | Bug: [test_failure_description]",
-          "            Minimal surgical fixes only - stay in Green phase",
-          "     \" > green-fix-iteration-[N]-diagnosis.md)",
-          "  2. Apply Fix (check meta.use_codex):",
-          "     IF meta.use_codex=false (default): Present diagnosis to user for manual fix",
-          "     IF meta.use_codex=true: Codex applies fix automatically",
-          "  3. Retest: bash(npm test -- tests/auth/login.test.ts)",
-          "  4. If pass → Exit loop, complete task",
-          "     If fail → Continue to next iteration",
-          "IF max_iterations reached: Revert changes, report failure"
-        ]
+    "implementation_approach": [
+      {
+        "step": 1,
+        "title": "Implement minimal code to pass tests",
+        "description": "Write minimal code based on test requirements following TDD principles - no over-engineering",
+        "modification_points": [
+          "Load test requirements from TEST phase",
+          "Create/modify implementation files",
+          "Implement only what tests require",
+          "Focus on passing tests, not perfection"
+        ],
+        "logic_flow": [
+          "Load test requirements from [test_requirements]",
+          "Parse test expectations and edge cases",
+          "Write minimal implementation code",
+          "Avoid premature optimization or abstraction"
+        ],
+        "depends_on": [],
+        "output": "initial_implementation"
+      },
+      {
+        "step": 2,
+        "title": "Test and iteratively fix until passing",
+        "description": "Run tests and enter iterative fix cycle if needed (max 3 iterations with auto-revert on failure)",
+        "modification_points": [
+          "Execute test suite",
+          "If tests fail: diagnose with Gemini",
+          "Apply fixes (manual or Codex if meta.use_codex=true)",
+          "Retest and iterate"
+        ],
+        "logic_flow": [
+          "Run test suite",
+          "If all tests pass → Complete",
+          "If tests fail → Enter iteration loop (max 3):",
+          "  Extract failure messages and stack traces",
+          "  Use Gemini bug-fix template for diagnosis",
+          "  Generate targeted fix recommendations",
+          "  Apply fixes (manual or Codex)",
+          "  Rerun tests",
+          "  If pass → Complete, if fail → Continue iteration",
+          "If max_iterations reached → Trigger auto-revert"
+        ],
+        "command": "bash(npm test -- tests/auth/login.test.ts)",
+        "depends_on": [1],
+        "output": "test_results"
      }
-    },
+    ],
    "post_completion": [
      {
        "step": "verify_tests_passing",
--- a/.claude/commands/workflow/tools/task-generate.md
+++ b/.claude/commands/workflow/tools/task-generate.md
@@ -1,9 +1,10 @@
 ---
 name: task-generate
 description: Generate task JSON files and IMPL_PLAN.md from analysis results with artifacts integration
-argument-hint: "--session WFS-session-id"
+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
@@ -11,12 +12,30 @@ examples:
 ## Overview
 Generate task JSON files and IMPL_PLAN.md from analysis results with automatic artifact detection and integration.

+## Execution Modes
+
+### Agent Mode (Default)
+Tasks execute within agent context using agent's capabilities:
+- Agent reads synthesis specifications
+- Agent implements following requirements
+- Agent validates implementation
+- **Benefit**: Seamless context within single agent execution
+
+### CLI Execute Mode (`--cli-execute`)
+Tasks execute using Codex CLI with resume mechanism:
+- Each task uses `codex exec` command in `implementation_approach`
+- First task establishes Codex session
+- Subsequent tasks use `codex exec "..." resume --last` for context continuity
+- **Benefit**: Codex's autonomous development capabilities with persistent context
+- **Use Case**: Complex implementation requiring Codex's reasoning and iteration
+
 ## Core Philosophy
 - **Analysis-Driven**: Generate from ANALYSIS_RESULTS.md
 - **Artifact-Aware**: Auto-detect brainstorming outputs
 - **Context-Rich**: Embed comprehensive context in task JSON
 - **Flow-Control Ready**: Pre-define implementation steps
 - **Memory-First**: Reuse loaded documents from memory
+- **CLI-Aware**: Support Codex resume mechanism for persistent context

 ## Core Responsibilities
 - Parse analysis results and extract tasks
@@ -153,23 +172,52 @@ Generate task JSON files and IMPL_PLAN.md from analysis results with automatic a
        "on_error": "fail"
      }
    ],
-    "implementation_approach": {
-      "task_description": "Implement '[title]' following synthesis specification. PRIORITY: Use synthesis-specification.md 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 synthesis-specification.md",
-        "Follow technical guidelines from synthesis",
-        "Consult artifacts for implementation details when needed",
-        "Integrate with existing patterns"
-      ],
-      "logic_flow": [
-        "Load synthesis specification",
-        "Extract requirements and design",
-        "Analyze existing patterns",
-        "Implement following specification",
-        "Consult artifacts for technical details when needed",
-        "Validate against acceptance criteria"
-      ]
-    },
+    "implementation_approach": [
+      {
+        "step": 1,
+        "title": "Implement task following synthesis specification",
+        "description": "Implement '[title]' following synthesis specification. PRIORITY: Use synthesis-specification.md 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 synthesis-specification.md",
+          "Follow technical guidelines from synthesis",
+          "Consult artifacts for implementation details when needed",
+          "Integrate with existing patterns"
+        ],
+        "logic_flow": [
+          "Load synthesis specification",
+          "Extract requirements and design",
+          "Analyze existing patterns",
+          "Implement following specification",
+          "Consult artifacts for technical details when needed",
+          "Validate against acceptance criteria"
+        ],
+        "depends_on": [],
+        "output": "implementation"
+      }
+    ],
+
+    // CLI Execute Mode: Use Codex command (when --cli-execute flag present)
+    "implementation_approach": [
+      {
+        "step": 1,
+        "title": "Execute implementation with Codex",
+        "description": "Use Codex CLI to implement '[title]' following synthesis specification with autonomous development capabilities",
+        "modification_points": [
+          "Codex loads synthesis specification and artifacts",
+          "Codex implements following requirements",
+          "Codex validates and tests implementation"
+        ],
+        "logic_flow": [
+          "Establish or resume Codex session",
+          "Pass synthesis specification to Codex",
+          "Codex performs autonomous implementation",
+          "Codex validates against acceptance criteria"
+        ],
+        "command": "bash(codex -C [focus_paths] --full-auto exec \"PURPOSE: [title] TASK: [requirements] MODE: auto CONTEXT: @{[synthesis_path],[artifacts_paths]} EXPECTED: [acceptance] RULES: Follow synthesis-specification.md\" [resume_flag] --skip-git-repo-check -s danger-full-access)",
+        "depends_on": [],
+        "output": "implementation"
+      }
+    ],
    "target_files": ["file:function:lines"]
  }
 }
@@ -181,7 +229,37 @@ Generate task JSON files and IMPL_PLAN.md from analysis results with automatic a
 3. Generate task context (requirements, focus_paths, acceptance)
 4. **Determine modification targets**: Extract specific code locations from analysis
 5. Build flow_control with artifact loading steps and target_files
-6. Create individual task JSON files in `.task/`
+6. **CLI Execute Mode**: If `--cli-execute` flag present, generate Codex commands
+7. Create individual task JSON files in `.task/`
+
+#### Codex Resume Mechanism (CLI Execute Mode)
+
+**Session Continuity Strategy**:
+- **First Task** (no depends_on or depends_on=[]): Establish new Codex session
+  - Command: `codex -C [path] --full-auto exec "[prompt]" --skip-git-repo-check -s danger-full-access`
+  - Creates new session context
+
+- **Subsequent Tasks** (has depends_on): Resume previous Codex session
+  - Command: `codex --full-auto exec "[prompt]" resume --last --skip-git-repo-check -s danger-full-access`
+  - Maintains context from previous implementation
+  - **Critical**: `resume --last` flag enables context continuity
+
+**Resume Flag Logic**:
+```javascript
+// Determine resume flag based on task dependencies
+const resumeFlag = task.context.depends_on && task.context.depends_on.length > 0
+  ? "resume --last"
+  : "";
+
+// First task (IMPL-001): no resume flag
+// Later tasks (IMPL-002, IMPL-003): use "resume --last"
+```
+
+**Benefits**:
+- ✅ Shared context across related tasks
+- ✅ Codex learns from previous implementations
+- ✅ Consistent patterns and conventions
+- ✅ Reduced redundant analysis

 #### Target Files Generation (Critical)
 **Purpose**: Identify specific code locations for modification AND new files to create
@@ -542,8 +620,82 @@ Core requirements, objectives, technical approach summary (2-3 paragraphs max).
 /workflow:tools:task-generate --session WFS-auth
 ```

+## CLI Execute Mode Examples
+
+### Example 1: First Task (Establish Session)
+```json
+{
+  "id": "IMPL-001",
+  "title": "Implement user authentication module",
+  "context": {
+    "depends_on": [],
+    "focus_paths": ["src/auth"],
+    "requirements": ["JWT-based authentication", "Login and registration endpoints"]
+  },
+  "flow_control": {
+    "implementation_approach": [{
+      "step": 1,
+      "title": "Execute implementation with Codex",
+      "command": "bash(codex -C src/auth --full-auto exec \"PURPOSE: Implement user authentication module TASK: JWT-based authentication with login and registration MODE: auto CONTEXT: @{.workflow/WFS-session/.brainstorming/synthesis-specification.md} EXPECTED: Complete auth module with tests RULES: Follow synthesis specification\" --skip-git-repo-check -s danger-full-access)",
+      "depends_on": [],
+      "output": "implementation"
+    }]
+  }
+}
+```
+
+### Example 2: Subsequent Task (Resume Session)
+```json
+{
+  "id": "IMPL-002",
+  "title": "Add password reset functionality",
+  "context": {
+    "depends_on": ["IMPL-001"],
+    "focus_paths": ["src/auth"],
+    "requirements": ["Password reset via email", "Token validation"]
+  },
+  "flow_control": {
+    "implementation_approach": [{
+      "step": 1,
+      "title": "Execute implementation with Codex",
+      "command": "bash(codex --full-auto exec \"PURPOSE: Add password reset functionality TASK: Password reset via email with token validation MODE: auto CONTEXT: Previous auth implementation from session EXPECTED: Password reset endpoints with email integration RULES: Maintain consistency with existing auth patterns\" resume --last --skip-git-repo-check -s danger-full-access)",
+      "depends_on": [],
+      "output": "implementation"
+    }]
+  }
+}
+```
+
+### Example 3: Third Task (Continue Session)
+```json
+{
+  "id": "IMPL-003",
+  "title": "Implement role-based access control",
+  "context": {
+    "depends_on": ["IMPL-001", "IMPL-002"],
+    "focus_paths": ["src/auth"],
+    "requirements": ["User roles and permissions", "Middleware for route protection"]
+  },
+  "flow_control": {
+    "implementation_approach": [{
+      "step": 1,
+      "title": "Execute implementation with Codex",
+      "command": "bash(codex --full-auto exec \"PURPOSE: Implement role-based access control TASK: User roles, permissions, and route protection middleware MODE: auto CONTEXT: Existing auth system from session EXPECTED: RBAC system integrated with current auth RULES: Use established patterns from session context\" resume --last --skip-git-repo-check -s danger-full-access)",
+      "depends_on": [],
+      "output": "implementation"
+    }]
+  }
+}
+```
+
+**Pattern Summary**:
+- IMPL-001: Fresh start with `-C src/auth` and full prompt
+- IMPL-002: Resume with `resume --last`, references "previous auth implementation"
+- IMPL-003: Resume with `resume --last`, references "existing auth system"
+
 ## 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:concept-enhanced` - Provides analysis results
 - `/workflow:execute` - Executes generated tasks
--- a/.claude/commands/workflow/tools/test-task-generate.md
+++ b/.claude/commands/workflow/tools/test-task-generate.md
@@ -1,10 +1,12 @@
 ---
 name: test-task-generate
 description: Generate test-fix task JSON with iterative test-fix-retest cycle specification
-argument-hint: "[--use-codex] --session WFS-test-session-id"
+argument-hint: "[--use-codex] [--cli-execute] --session WFS-test-session-id"
 examples:
  - /workflow:tools:test-task-generate --session WFS-test-auth
  - /workflow:tools:test-task-generate --use-codex --session WFS-test-auth
+  - /workflow:tools:test-task-generate --cli-execute --session WFS-test-auth
+  - /workflow:tools:test-task-generate --cli-execute --use-codex --session WFS-test-auth
 ---

 # Test Task Generation Command
@@ -12,6 +14,16 @@ examples:
 ## Overview
 Generate specialized test-fix task JSON with comprehensive test-fix-retest cycle specification, including Gemini diagnosis (using bug-fix template) and manual fix workflow (Codex automation only when explicitly requested).

+## Execution Modes
+
+### Test Generation (IMPL-001)
+- **Agent Mode (Default)**: @code-developer generates tests within agent context
+- **CLI Execute Mode (`--cli-execute`)**: Use Codex CLI for autonomous test generation
+
+### Test Fix (IMPL-002)
+- **Manual Mode (Default)**: Gemini diagnosis → user applies fixes
+- **Codex Mode (`--use-codex`)**: Gemini diagnosis → Codex applies fixes with resume mechanism
+
 ## Core Philosophy
 - **Analysis-Driven Test Generation**: Use TEST_ANALYSIS_RESULTS.md from test-concept-enhanced
 - **Agent-Based Test Creation**: Call @code-developer agent for comprehensive test generation
@@ -37,8 +49,9 @@ Generate specialized test-fix task JSON with comprehensive test-fix-retest cycle
 ### Phase 1: Input Validation & Discovery

 1. **Parameter Parsing**
-   - Parse `--use-codex` flag from command arguments
-   - Store flag value for IMPL-002.json generation
+   - Parse `--use-codex` flag from command arguments → Controls IMPL-002 fix mode
+   - Parse `--cli-execute` flag from command arguments → Controls IMPL-001 generation mode
+   - Store flag values for task JSON generation

 2. **Test Session Validation**
   - Load `.workflow/{test-session-id}/workflow-session.json`
@@ -142,31 +155,53 @@ Generate **TWO task JSON files**:
        "on_error": "skip_optional"
      }
    ],
-    "implementation_approach": {
-      "task_description": "Generate comprehensive test suite based on TEST_ANALYSIS_RESULTS.md. Follow test generation strategy and create all test files listed in section 5 (Implementation Targets).",
-      "generation_steps": [
-        "Read TEST_ANALYSIS_RESULTS.md section 3 (Test Requirements by File)",
-        "Read TEST_ANALYSIS_RESULTS.md section 4 (Test Generation Strategy)",
-        "Study existing test patterns from test_context.test_framework.conventions",
-        "For each test file in section 5 (Implementation Targets):",
-        "  - Create test file with specified scenarios",
-        "  - Implement happy path tests",
-        "  - Implement error handling tests",
-        "  - Implement edge case tests",
-        "  - Implement integration tests (if specified)",
-        "  - Add required mocks and fixtures",
-        "Follow test framework conventions and project standards",
-        "Ensure all tests are executable and syntactically valid"
+    // Agent Mode (Default): Agent implements tests
+    "implementation_approach": [
+      {
+        "step": 1,
+        "title": "Generate comprehensive test suite",
+        "description": "Generate comprehensive test suite based on TEST_ANALYSIS_RESULTS.md. Follow test generation strategy and create all test files listed in section 5 (Implementation Targets).",
+        "modification_points": [
+          "Read TEST_ANALYSIS_RESULTS.md sections 3 and 4",
+          "Study existing test patterns",
+          "Create test files with all required scenarios",
+          "Implement happy path, error handling, edge case, and integration tests",
+          "Add required mocks and fixtures"
+        ],
+        "logic_flow": [
+          "Read TEST_ANALYSIS_RESULTS.md section 3 (Test Requirements by File)",
+          "Read TEST_ANALYSIS_RESULTS.md section 4 (Test Generation Strategy)",
+          "Study existing test patterns from test_context.test_framework.conventions",
+          "For each test file in section 5 (Implementation Targets): Create test file with specified scenarios, Implement happy path tests, Implement error handling tests, Implement edge case tests, Implement integration tests (if specified), Add required mocks and fixtures",
+          "Follow test framework conventions and project standards",
+          "Ensure all tests are executable and syntactically valid"
+        ],
+        "depends_on": [],
+        "output": "test_suite"
+      }
+    ],
+
+    // CLI Execute Mode (--cli-execute): Use Codex command (alternative format shown below)
+    "implementation_approach": [{
+      "step": 1,
+      "title": "Generate tests using Codex",
+      "description": "Use Codex CLI to autonomously generate comprehensive test suite based on TEST_ANALYSIS_RESULTS.md",
+      "modification_points": [
+        "Codex loads TEST_ANALYSIS_RESULTS.md and existing test patterns",
+        "Codex generates all test files listed in analysis section 5",
+        "Codex ensures tests follow framework conventions"
      ],
-      "quality_criteria": [
-        "All test scenarios from analysis are implemented",
-        "Test structure matches existing patterns",
-        "Clear test descriptions and assertions",
-        "Proper setup/teardown and fixtures",
-        "Dependencies properly mocked",
-        "Tests follow project coding standards"
-      ]
-    },
+      "logic_flow": [
+        "Start new Codex session",
+        "Pass TEST_ANALYSIS_RESULTS.md to Codex",
+        "Codex studies existing test patterns",
+        "Codex generates comprehensive test suite",
+        "Codex validates test syntax and executability"
+      ],
+      "command": "bash(codex -C [focus_paths] --full-auto exec \"PURPOSE: Generate comprehensive test suite TASK: Create test files based on TEST_ANALYSIS_RESULTS.md section 5 MODE: write CONTEXT: @{.workflow/WFS-test-[session]/.process/TEST_ANALYSIS_RESULTS.md,.workflow/WFS-test-[session]/.process/test-context-package.json} EXPECTED: All test files with happy path, error handling, edge cases, integration tests RULES: Follow test framework conventions, ensure tests are executable\" --skip-git-repo-check -s danger-full-access)",
+      "depends_on": [],
+      "output": "test_generation"
+    }],
    "target_files": [
      "{test_file_1 from TEST_ANALYSIS_RESULTS.md section 5}",
      "{test_file_2 from TEST_ANALYSIS_RESULTS.md section 5}",
@@ -278,24 +313,27 @@ Generate **TWO task JSON files**:
        "on_error": "skip_optional"
      }
    ],
-    "implementation_approach": {
-      "task_description": "Execute iterative test-fix-retest cycle using Gemini diagnosis (bug-fix template) and manual fixes (Codex only if explicitly needed)",
-      "test_fix_cycle": {
-        "max_iterations": 5,
-        "cycle_pattern": "test → gemini_diagnose → manual_fix (or codex if needed) → retest",
-        "tools": {
-          "test_execution": "bash(test_command)",
-          "diagnosis": "gemini-wrapper (MODE: analysis, uses bug-fix template)",
-          "fix_application": "manual (default) or codex exec resume --last (if explicitly needed)",
-          "verification": "bash(test_command) + regression_check"
+    "implementation_approach": [
+      {
+        "step": 1,
+        "title": "Execute iterative test-fix-retest cycle",
+        "description": "Execute iterative test-fix-retest cycle using Gemini diagnosis (bug-fix template) and manual fixes (Codex only if meta.use_codex=true). Max 5 iterations with automatic revert on failure.",
+        "test_fix_cycle": {
+          "max_iterations": 5,
+          "cycle_pattern": "test → gemini_diagnose → manual_fix (or codex if needed) → retest",
+          "tools": {
+            "test_execution": "bash(test_command)",
+            "diagnosis": "gemini-wrapper (MODE: analysis, uses bug-fix template)",
+            "fix_application": "manual (default) or codex exec resume --last (if explicitly needed)",
+            "verification": "bash(test_command) + regression_check"
+          },
+          "exit_conditions": {
+            "success": "all_tests_pass",
+            "failure": "max_iterations_reached",
+            "error": "test_command_not_found"
+          }
        },
-        "exit_conditions": {
-          "success": "all_tests_pass",
-          "failure": "max_iterations_reached",
-          "error": "test_command_not_found"
-        }
-      },
-      "modification_points": [
+        "modification_points": [
        "PHASE 1: Initial Test Execution",
        "  1.1. Discover test command from framework detection",
        "  1.2. Execute initial test run: bash([test_command])",
@@ -427,33 +465,36 @@ Generate **TWO task JSON files**:
        "  Generate summary (include test generation info)",
        "  Certify code APPROVED"
      ],
-      "error_handling": {
-        "max_iterations_reached": {
-          "action": "revert_all_changes",
-          "commands": [
-            "bash(git reset --hard HEAD)",
-            "bash(jq '.status = \"blocked\"' .workflow/[session]/.task/IMPL-001.json > temp.json && mv temp.json .workflow/[session]/.task/IMPL-001.json)"
-          ],
-          "report": "Generate failure report with iteration logs in .summaries/IMPL-001-failure-report.md"
+        "error_handling": {
+          "max_iterations_reached": {
+            "action": "revert_all_changes",
+            "commands": [
+              "bash(git reset --hard HEAD)",
+              "bash(jq '.status = \"blocked\"' .workflow/[session]/.task/IMPL-001.json > temp.json && mv temp.json .workflow/[session]/.task/IMPL-001.json)"
+            ],
+            "report": "Generate failure report with iteration logs in .summaries/IMPL-001-failure-report.md"
+          },
+          "test_command_fails": {
+            "action": "treat_as_test_failure",
+            "context": "Use stderr as failure context for Gemini diagnosis"
+          },
+          "codex_apply_fails": {
+            "action": "retry_once_then_skip",
+            "fallback": "Mark iteration as skipped, continue to next"
+          },
+          "gemini_diagnosis_fails": {
+            "action": "retry_with_simplified_context",
+            "fallback": "Use previous diagnosis, continue"
+          },
+          "regression_detected": {
+            "action": "log_warning_continue",
+            "context": "Include regression info in next Gemini diagnosis"
+          }
        },
-        "test_command_fails": {
-          "action": "treat_as_test_failure",
-          "context": "Use stderr as failure context for Gemini diagnosis"
-        },
-        "codex_apply_fails": {
-          "action": "retry_once_then_skip",
-          "fallback": "Mark iteration as skipped, continue to next"
-        },
-        "gemini_diagnosis_fails": {
-          "action": "retry_with_simplified_context",
-          "fallback": "Use previous diagnosis, continue"
-        },
-        "regression_detected": {
-          "action": "log_warning_continue",
-          "context": "Include regression info in next Gemini diagnosis"
-        }
+        "depends_on": [],
+        "output": "test_fix_results"
      }
-    },
+    ],
    "target_files": [
      "Auto-discovered from test failures",
      "Extracted from Gemini diagnosis each iteration",