메뉴
Code research activity skill for PAW workflow. Documents implementation details with file:line references, discovers documentation infrastructure, and creates CodeResearch.md artifact.
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/lossyrob/phased-agent-workflow --skill paw-code-research명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
SKILL.md 표시 중
SOC 직업 분류 기준
| name | paw-code-research |
| description | Code research activity skill for PAW workflow. Documents implementation details with file:line references, discovers documentation infrastructure, and creates CodeResearch.md artifact. |
Execution Context: This skill runs in a subagent session, delegated by the PAW orchestrator. Return structured results to the orchestrator upon completion.
Document where and how code works with precise file:line references. Creates a technical map of the existing system for implementation planning.
Reference: Follow Core Implementation Principles from
paw-workflowskill.
Document:
Your sole purpose is to document what exists. You are a technical cartographer mapping existing terrain, not a consultant evaluating it.
NEVER:
ONLY describe what exists, where it exists, and how components interact.
May document neutrally (with file:line evidence): observed constraints, limitations, or risks that inform planning decisions—but without proposing solutions.
Relationship to SpecResearch.md: If SpecResearch.md exists, build on that behavioral understanding. If not, focus on implementation details relevant to Spec.md requirements.
Guidelines: Include file:line references for all claims. Read thoroughly before stating. Trace actual paths—don't assume.
Research documentation infrastructure as a standard component. Capture in CodeResearch.md:
| Aspect | What to Find |
|---|---|
| Framework | mkdocs, docusaurus, sphinx, plain markdown, none |
| Docs Directory | Path to docs folder (e.g., docs/, documentation/) |
| Navigation Config | mkdocs.yml, sidebar.js, conf.py, etc. |
| Style Conventions | Verbosity level, heading patterns, code block usage |
| Build Command | Command to build/serve docs locally |
| Standard Files | README.md, CHANGELOG.md, CONTRIBUTING.md locations |
Why: Planning skill uses this to determine if documentation phases are warranted and how to structure them.
Research project verification commands. Capture in CodeResearch.md:
| Aspect | What to Find |
|---|---|
| Test Command | npm test, make test, pytest, etc. |
| Lint Command | npm run lint, make lint, eslint, etc. |
| Build Command | npm run build, make build, tsc, etc. |
| Type Check | npm run typecheck, tsc --noEmit, mypy, etc. |
Why: Implementation skill uses these to verify changes before committing.
Save to: .paw/work/<work-id>/CodeResearch.md
---
date: [ISO timestamp with timezone]
git_commit: [commit hash]
branch: [branch name]
repository: [repo name]
topic: "[Research topic]"
tags: [research, codebase, component-names]
status: complete
last_updated: [YYYY-MM-DD]
---
# Research: [Topic]
## Research Question
[Original query or derived from Spec.md]
## Summary
[High-level findings answering the research question]
## Documentation System
- **Framework**: [mkdocs/docusaurus/sphinx/markdown/none]
- **Docs Directory**: [path or N/A]
- **Navigation Config**: [path or N/A]
- **Style Conventions**: [key observations]
- **Build Command**: [command or N/A]
- **Standard Files**: [README, CHANGELOG locations]
## Verification Commands
- **Test Command**: [e.g., `npm test`, `make test`, `pytest`]
- **Lint Command**: [e.g., `npm run lint`, `make lint`]
- **Build Command**: [e.g., `npm run build`, `make build`]
- **Type Check**: [e.g., `npm run typecheck`, `tsc --noEmit`]
## Detailed Findings
### [Component/Area 1]
- Description (`file.ext:line`, include permalink if available)
- How it connects to other components
- Current implementation details
### [Component/Area 2]
...
## Code References
- `path/to/file.py:123` - Description
- `another/file.ts:45-67` - Description
## Architecture Documentation
[Patterns, conventions, and design implementations found]
## Open Questions
[Areas needing further investigation, if any]
When on main branch or pushed commit, permalinks enhance traceability:
https://github.com/{owner}/{repo}/blob/{commit}/{file}#L{line}
Permalinks are optional—file:line references are the primary evidence requirement. If remote URL and commit SHA are readily available, include permalinks; otherwise proceed without them.
| Mode | Spec Artifact | Approach |
|---|---|---|
| full | Spec.md exists | Read Spec.md for context, comprehensive research |
| minimal | Spec.md may not exist | Use Issue URL as requirements source |
| custom | Check Custom Workflow Instructions | Adapt based on instructions |
If Spec.md expected but missing, note it and use Issue URL as fallback.
For additional questions after initial research:
## Follow-up Research [timestamp] sectionlast_updated, add last_updated_noteReport to PAW agent:
.paw/work/<work-id>/CodeResearch.mdTransforms evaluation findings into structured review comments with comprehensive rationale. Handles both initial draft generation and critique response iteration.
Posts finalized review comments to GitHub, leaving the review pending by default or submitting it under explicit verified authorization.
Analyzes PR changes to create ReviewContext.md and DerivedSpec.md artifacts. Handles both initial analysis and resumption after baseline research.