agentleFS
suitedacessuitedaces/dorabot/skills/review-pr/SKILL.md

review-pr

Skill247 starsChanged 53 days agoMITView on GitHub

Review GitHub pull requests with structured code analysis. Use when asked to review a PR, check a pull request, or audit code changes.

185 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use diagnose-ci-failures, docs-loop and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • zerx-labSkill

    diagnose-ci-failures

    Diagnose CI failures for a PR using the GitHub CLI, extract error logs, and generate a plan to fix them. Use when the user asks to check CI status, pull CI issues, triage test failures, or investigate PR build failures.

    Also in 3 projects that use this

  • GeiserXSkill

    docs-loop

    Use when the user asks to audit docs, check documentation for stale claims, or verify docs against the code.

    Also in 3 projects that use this

  • GeiserXSkill

    full-loop

    Drive a repository to done with sergio-loop, then keep improving it with refine-loop at a descending theta. Use only when the user explicitly invokes full-loop.

    Also in 3 projects that use this

  • n24q02mSkill

    review-delta

    Review only changes since last commit using impact analysis. Token-efficient delta review with automatic blast-radius detection.

    Also in 3 projects that use this

  • n24q02mSkill

    build-graph

    Build or update the code review knowledge graph. Run this first to initialize, or let hooks keep it updated automatically.

    Also in 3 projects that use this

  • +7 more from GeiserX

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 10 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 178 edited versions we read keep these.

  • 10

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If ambiguous, ask for the repo.

    line 19
  2. Distinctive:

    Run these in parallel:

    line 23
  3. Distinctive:

    Read the full diff carefully.

    line 44
  4. Distinctive:

    Don't nitpick formatting if there's a formatter configured.

    line 113
  5. Distinctive:

    If the PR is large (>500 lines), focus on the riskiest files first.

    line 114
  6. Distinctive:

    If the PR description explains a deliberate tradeoff, don't re-litigate it.

    line 115
  7. Distinctive:

    Check the test coverage: are new code paths tested?

    line 116
  8. Distinctive:

    Only post to GitHub if the user explicitly asks.

    line 120
  9. Distinctive:

    Always show the review to the user first and get confirmation before posting to GitHub.

    line 142
  10. Distinctive:

    When reviewing multiple PRs (e.g., "review all open PRs"):

    line 150

“Common” is measured: kept by at least 6 in 10 of the 178 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

PR Review Skill

Review GitHub pull requests with structured analysis covering correctness, security, performance, and style.

Invocation

Triggered by: /review-pr <url-or-number>, "review this PR", "review PR #123", or a GitHub PR URL.

Parse the input:

  • Full URL: https://github.com/owner/repo/pull/123 -> extract owner, repo, number
  • Number only: #123 or 123 -> use current repo from gh repo view --json nameWithOwner -q .nameWithOwner
  • If ambiguous, ask for the repo.

Step 1: Gather Context

Run these in parallel:

# PR metadata
gh pr view <number> --repo <owner/repo> --json title,body,author,baseRefName,headRefName,createdAt,additions,deletions,changedFiles,labels,reviewRequests,mergeable,state

# Full diff
gh pr diff <number> --repo <owner/repo>

# File list with stats
gh pr view <number> --repo <owner/repo> --json files --jq '.files[] | "\(.additions)+/\(.deletions)- \(.path)"'

# CI status
gh pr checks <number> --repo <owner/repo> 2>/dev/null || echo "No checks"

# Existing review comments (avoid duplicating feedback)
gh api repos/<owner>/<repo>/pulls/<number>/comments --jq '.[] | "[\(.path):\(.line // .original_line)] @\(.user.login): \(.body)"' 2>/dev/null | head -50

Step 2: Analyze the Diff

Read the full diff carefully. For each file, evaluate:

Correctness

  • Logic errors, off-by-one, null/undefined handling
  • Missing error handling or edge cases
  • Broken contracts (function signatures changed without updating callers)
  • Race conditions or concurrency issues

Security

  • User input flowing to SQL, shell commands, file paths, or HTML without sanitization
  • Hardcoded secrets, API keys, or credentials
  • Overly broad permissions or missing auth checks
  • Unsafe deserialization or eval usage

Performance

  • O(n^2) or worse algorithms where O(n) is possible
  • Unnecessary re-renders in React (missing memo/useMemo/useCallback)
  • N+1 queries or unbounded data fetching
  • Large allocations in hot paths

Architecture

  • Does this change belong in this file/module?
  • Are abstractions appropriate (not too much, not too little)?
  • Does it follow existing patterns in the codebase?
  • Will this be maintainable?

Style

  • Naming clarity
  • Dead code or TODOs left behind
  • Missing types (in TypeScript)
  • Inconsistent formatting (only flag if egregious)

Step 3: Write the Review

Output a structured review in this format:

## PR Review: <title>

**Summary**: 1-3 sentence overview of what this PR does and whether it's ready.

**Verdict**: APPROVE | REQUEST_CHANGES | COMMENT

### Findings

#### <severity-emoji> <category>: <short description>
**File**: `path/to/file.ts` L<line>-L<line>
<explanation of the issue and why it matters>

**Suggested fix**:
\`\`\`diff
- old code
+ new code
\`\`\`

---
(repeat for each finding)

Severity emojis:

  • [blocker] - Must fix before merge. Bugs, security issues, data loss risks.
  • [warning] - Should fix. Performance problems, poor patterns, missing edge cases.
  • [nit] - Optional. Style, naming, minor improvements.
  • [praise] - Good stuff. Call out well-written code (do this at least once).

Guidelines

  • Be specific. Quote the code, give line numbers, explain why.
  • Suggest fixes, not just problems. Include diff snippets.
  • Don't nitpick formatting if there's a formatter configured.
  • If the PR is large (>500 lines), focus on the riskiest files first.
  • If the PR description explains a deliberate tradeoff, don't re-litigate it.
  • Check the test coverage: are new code paths tested? Are edge cases covered?

Step 4: Post the Review (Optional)

Only post to GitHub if the user explicitly asks. Use:

# Post a review comment (not individual line comments)
gh pr review <number> --repo <owner/repo> --comment --body "<review body>"

# Or approve/request changes
gh pr review <number> --repo <owner/repo> --approve --body "<review body>"
gh pr review <number> --repo <owner/repo> --request-changes --body "<review body>"

For inline comments on specific lines:

gh api repos/<owner>/<repo>/pulls/<number>/comments \
  -f body="<comment>" \
  -f path="<file>" \
  -f commit_id="$(gh pr view <number> --repo <owner/repo> --json headRefOid -q .headRefOid)" \
  -F line=<line_number> \
  -f side="RIGHT"

Always show the review to the user first and get confirmation before posting to GitHub.

Quick Mode

For small PRs (<100 lines changed), skip the full structure. Give a concise paragraph covering the key points and verdict.

Batch Review

When reviewing multiple PRs (e.g., "review all open PRs"):

gh pr list --repo <owner/repo> --json number,title,author,additions,deletions --jq '.[] | "#\(.number) \(.title) by @\(.author.login) (+\(.additions)/-\(.deletions))"'

Triage by size and risk, review the largest/riskiest first.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.