- Implemented tests for the QueryParser class, covering various identifier splitting methods (CamelCase, snake_case, kebab-case), OR expansion, and FTS5 operator preservation.
- Added parameterized tests to validate expected token outputs for different query formats.
- Created edge case tests to ensure robustness against unusual input scenarios.
- Developed tests for the Reciprocal Rank Fusion (RRF) algorithm, including score computation, weight handling, and result ranking across multiple sources.
- Included tests for normalization of BM25 scores and tagging search results with source metadata.
- 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.
**Core Changes**:
- Shift from heavy analysis in planning to targeted analysis in execution
- Reduce main process context by collecting only metadata (paths, structure)
- Simplify pre_analysis steps to be more declarative
**docs.md optimizations**:
- Phase 2: Lightweight metadata collection (get_modules_by_depth.sh + Code Index MCP)
- Planning phase now collects only paths and structure, not file content
- IMPL-001: Simplified to 2 pre_analysis steps (structure + tech stack from config files only)
- IMPL-002: Removed system_context loading, single focused analysis step
- IMPL-N (API): Replace rg with mcp__code-index__search_code_advanced for better structure
**doc-generator.md optimizations**:
- Add "Scope-Limited Analysis" philosophy
- Add "Optimized Execution Model" section
- Emphasize analysis is always limited to focus_paths to prevent context explosion
**Benefits**:
- Main process context stays minimal (no risk of overflow)
- Each task has controlled, scoped context
- Better separation: planning = metadata, execution = deep analysis
- More maintainable pre_analysis steps (less nested commands)
- Better use of MCP tools for structured data
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Update doc-generator agent description to emphasize intelligent autonomous execution
- Add on_error field to all flow_control pre_analysis steps
- Clarify agent role as goal-oriented worker vs script runner
- Improve error handling documentation in workflow examples
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Updated `/workflow:docs` command to enhance documentation planning and orchestration.
- Introduced new parameters for tool selection and scope filtering.
- Separated documentation planning from content generation, clarifying roles of docs.md and doc-generator.md.
- Added detailed planning process phases, including session initialization, project structure analysis, and task decomposition.
- Created new templates for API, module, and project-level documentation, ensuring comprehensive coverage and consistency.
- Enhanced output structure for better organization of generated documentation.
- Implemented error handling and completion checklist for improved user experience.
- Add pre-analysis documentation check as first step
- Reference specific paths from /workflow:docs output structure
- Update flow_control to selectively load relevant documentation
- Remove outdated plan-deep command
- Add new doc-generator agent and workflow files
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
