feat: enhance gh-pr-review skill with multi-agent teams review and auto-fix#13530
Merged
feat: enhance gh-pr-review skill with multi-agent teams review and auto-fix#13530
Conversation
…to-fix Rewrite the gh-pr-review skill based on Tencent/tgfx's code review system. Add structured review checklists, adversarial reviewer-verifier mechanism, risk-based auto-fix, and self-diagnosis capability. New reference files: - teams-review.md: multi-agent parallel review with Agent tool - local-review.md: single-agent local change review - pr-review.md: PR review via gh api inline comments - diagnosis.md: skill self-diagnosis (/gh-pr-review diag) - code-checklist.md: adapted for TS/React/Electron stack - doc-checklist.md: document review checklist - judgment-matrix.md: risk levels with v2 block rules - checklist-evolution.md: checklist update mechanism Co-Authored-By: Claude Opus 4.6 <[email protected]> Signed-off-by: suyao <[email protected]>
Add cross-references to the vercel-react-best-practices skill from code-checklist.md, teams-review.md, local-review.md, and pr-review.md for deeper React/performance analysis during reviews. Co-Authored-By: Claude Opus 4.6 <[email protected]> Signed-off-by: suyao <[email protected]>
Remove outdated reference to gh-pr-review CLI extension in the skill frontmatter description. Co-Authored-By: Claude Opus 4.6 <[email protected]> Signed-off-by: suyao <[email protected]>
Replace raw gh api calls with gh-pr-review CLI extension (EurFelux/gh-pr-review) for structured pending reviews with inline comments, preview, and submit. Co-Authored-By: Claude Opus 4.6 <[email protected]> Signed-off-by: suyao <[email protected]>
GeorgeDong32
approved these changes
Mar 17, 2026
Collaborator
GeorgeDong32
left a comment
There was a problem hiding this comment.
Summary
This is a well-designed enhancement to the gh-pr-review skill, inspired by Tencent/tgfx's battle-tested code review system. The multi-agent teams review with adversarial verification (reviewer vs verifier) is a solid approach to reduce false positives.
Highlights
- Clean routing logic in
SKILL.md- routes to appropriate review flow based on context - Comprehensive checklists tailored for the project's TypeScript/React/Electron stack
- Risk-based auto-fix with clear judgment matrix for fix decisions
- V2 block rules properly documented to prevent conflicts with ongoing refactoring
- Cross-references to
vercel-react-best-practicesfor React/performance deep checks
Minor Observations (no action needed)
-
The
git commit --onlyflag inteams-review.mdFix rules is widely supported since Git 2.0.5+ (2014), so no compatibility concerns. -
The module partition thresholds (≤1000 lines, ≤20 files) for single-reviewer optimization seem reasonable for this codebase.
Verdict: Ready to merge. Good documentation quality and clear architectural decisions.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What this PR does
Before this PR:
The
gh-pr-reviewskill was a simple single-agent PR review workflow using thegh-pr-reviewCLI extension (EurFelux/gh-pr-review) for structured inline comments. It only supported PR reviews.After this PR:
The
gh-pr-reviewskill is a comprehensive code review system inspired by Tencent/tgfx's.codebuddy/skills/cr, adapted for Claude Code's Agent tool. It now supports:vercel-react-best-practicesfor deeper React/performance analysis/gh-pr-review diagto analyze and improve the skill itselfgh-pr-reviewCLI extension for pending reviews with inline comments, preview, and structured submissionWhy we need it and why it was done in this way
The previous skill was limited to PR-only review. The tgfx approach provides a battle-tested, comprehensive review system with adversarial verification (reviewer finds issues, verifier challenges them), which significantly reduces false positives.
The following tradeoffs were made:
CreateTeam/SendMessage/TeamDeleteAPIs to Claude Code'sAgenttool with parallel subagentsvercel-react-best-practicesskill for React/performance checks instead of duplicating those rulesgh-pr-reviewCLI extension for PR review comment submission (pending review API with preview)The following alternatives were considered:
/crskill alongsidegh-pr-review— rejected because both serve code review and having two review skills would be confusingLinks to places where the discussion took place:
Breaking changes
None. The skill name and invocation (
/gh-pr-review) remain the same. Thegh-pr-reviewCLI extension is still used for PR reviews. New capabilities (local review, teams review, auto-fix) are additive.Special notes for your reviewer
references/teams-review.mdis the most complex file — it orchestrates multi-agent review with 6 phasesvercel-react-best-practicesskill for React/performance deep checksChecklist
/gh-pr-review,gh pr diff, or GitHub UI) before requesting review from othersRelease note