一键导入
requesting-code-review
Use when completing tasks, implementing major features, or before merging to verify work meets requirements
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Use when completing tasks, implementing major features, or before merging to verify work meets requirements
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Discover FRESH issues in WELL-MAINTAINED repos (200+ stars). Merge-optimized: 60% easy wins (docs, typos, tests) + 40% bug fixes. Target agentic AI repos by CRITERIA (topic:llm/agent/rag + stars:>200). Verify repo health before queuing.
Orchestrator-level PR follow-up detection and delegation: scan open PRs for new review comments, classify feedback type, write context files to subagent-inputs, delegate to oss-pr-review-handler sub-agents. Does NOT implement changes directly.
Sub-agent skill for handling PR review feedback: clone repo, checkout PR branch, read all review comments (inline + general), implement requested changes, push updates, respond to reviewers via gh CLI, cleanup workspace. Spawned 1-per-PR by the orchestrator.
Submit a contribution PR (bug fix, docs fix, typo fix, or test addition) to an open-source repo: verify it's a valid contribution, push branch to fork, create PR with gh CLI, use repo's PR template, log submission, report to dashboard.
Triage GitHub issues: confirm contribution type (bug/docs/typo/test), assess feasibility, check repo health (merge velocity, review rate). Merge-optimized scoring: +5 docs/typo, +3 tests, +5 fast merge repos. REJECT issues in unhealthy repos.
Analyze repo health and conventions BEFORE queuing any issue. MANDATORY health gate: stars >=200, last commit <2wk, merge time <14d, review rate >50%, open PRs <50. Use /Users/kevinlin/clawOSS/scripts/repo-health-check.sh for automation.
| name | requesting-code-review |
| description | Use when completing tasks, implementing major features, or before merging to verify work meets requirements |
Review your own diff using the checklist below to catch issues before they cascade. This keeps the review focused on the work product.
Core principle: Review early, review often.
Mandatory:
Optional but valuable:
1. Get git SHAs:
BASE_SHA=$(git rev-parse HEAD~1) # or origin/main
HEAD_SHA=$(git rev-parse HEAD)
2. Review the diff yourself using the checklist below:
Run git diff $BASE_SHA $HEAD_SHA and evaluate against:
3. Act on feedback:
[Just completed Task 2: Add verification function]
You: Let me review before proceeding.
BASE_SHA=$(git log --oneline | grep "Task 1" | head -1 | awk '{print $1}')
HEAD_SHA=$(git rev-parse HEAD)
git diff $BASE_SHA $HEAD_SHA
[Self-review findings]:
Strengths: Clean architecture, real tests
Issues:
Important: Missing progress indicators
Minor: Magic number (100) for reporting interval
Assessment: Ready to proceed after fixing Important issues
You: [Fix progress indicators]
[Continue to Task 3]
Subagent-Driven Development:
Executing Plans:
Ad-Hoc Development:
Never:
If reviewer wrong:
See template at: requesting-code-review/code-reviewer.md