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.
This commit is contained in:
catlog22
2025-10-12 14:09:47 +08:00
parent c5aac313ff
commit 58153ecb83
15 changed files with 1479 additions and 920 deletions

View File

@@ -142,37 +142,6 @@ mcp__exa__get_code_context_exa(
}
```
**Legacy Pre-Execution Analysis** (backward compatibility):
- **Multi-step Pre-Analysis**: Execute comprehensive analysis BEFORE implementation begins
- **Sequential Processing**: Process each step sequentially, expanding brief actions
- **Template Usage**: Use full template paths with $(cat template_path)
- **Method Selection**: gemini/codex/manual/auto-detected
- **CLI Commands**:
- **Gemini**: `bash(~/.claude/scripts/gemini-wrapper -p "$(cat template_path) [action]")`
- **Codex**: `bash(codex --full-auto exec "$(cat template_path) [action]" -s danger-full-access)`
- **Follow Guidelines**: @~/.claude/workflows/intelligent-tools-strategy.md
### Pre-Execution Analysis
**When [MULTI_STEP_ANALYSIS] marker is present:**
#### Multi-Step Pre-Analysis Execution
1. Process each analysis step sequentially from pre_analysis array
2. For each step:
- Expand brief action into comprehensive analysis task
- Use specified template with $(cat template_path)
- Execute with specified method (gemini/codex/manual/auto-detected)
3. Accumulate results across all steps for comprehensive context
4. Use consolidated analysis to inform implementation stages and task breakdown
#### Analysis Dimensions Coverage
- **Exa Research**: Use `mcp__exa__get_code_context_exa` for technology stack selection and API patterns
- Architecture patterns and component relationships
- Implementation conventions and coding standards
- Module dependencies and integration points
- Testing requirements and coverage patterns
- Security considerations and performance implications
3. Use Codex insights to create self-guided implementation stages
## Core Functions
### 1. Stage Design
@@ -222,11 +191,26 @@ Generate individual `.task/IMPL-*.json` files with:
"output_to": "codebase_structure"
}
],
"implementation_approach": {
"task_description": "Implement following synthesis specification",
"modification_points": ["Apply requirements"],
"logic_flow": ["Load spec", "Analyze", "Implement", "Validate"]
},
"implementation_approach": [
{
"step": 1,
"title": "Load and analyze synthesis specification",
"description": "Load synthesis specification from artifacts and extract requirements",
"modification_points": ["Load synthesis specification", "Extract requirements and design patterns"],
"logic_flow": ["Read synthesis specification from artifacts", "Parse architecture decisions", "Extract implementation requirements"],
"depends_on": [],
"output": "synthesis_requirements"
},
{
"step": 2,
"title": "Implement following specification",
"description": "Implement task requirements following consolidated synthesis specification",
"modification_points": ["Apply requirements from [synthesis_requirements]", "Modify target files", "Integrate with existing code"],
"logic_flow": ["Apply changes based on [synthesis_requirements]", "Implement core logic", "Validate against acceptance criteria"],
"depends_on": [1],
"output": "implementation"
}
],
"target_files": ["file:function:lines", "path/to/NewFile.ts"]
}
}