agentleFS
Sign inStart free
milad-ahmdmilad-ahmd/agent-skills/skills/pr-reviewer-pro/SKILL.md

pr-reviewer-pro

Skill5 starsChanged 18 days agoMITView on GitHub

Reviews a pull request or local branch diff like a senior engineer, covering correctness, security, performance, naming, and missing tests, and produces prioritized, actionable comments (optionally posted to GitHub). Use when the user asks to review a PR, a diff, a branch, or "check my changes before I push".

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use the helper script (works for both cases):

    line 12
  2. Very specific:

    Evidence or silence: Every comment cites file:line and explains *why* it matters.

    line 35
  3. Distinctive:

    If unsure, phrase it as a question rather than a claim.

    line 40
  4. Distinctive:

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

    line 69

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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 Reviewer Pro

Give a review that a senior engineer would be proud of: few, high-signal, actionable comments.

1. Collect the diff

Use the helper script (works for both cases):

bash scripts/collect_diff.sh            # local: current branch vs default branch
bash scripts/collect_diff.sh 123        # GitHub PR #123 (needs `gh` CLI)

Then read surrounding code for every changed hunk; never review a hunk in isolation. Also read the PR description / commit messages to understand the intent.

2. Review checklist (in priority order)

SeverityLook for
🔴 BlockerBugs, broken logic, data loss, race conditions, unhandled errors, security holes (injection, authz bypass, secrets in code, unsafe deserialization, SSRF, XSS), breaking API changes without migration
🟠 MajorN+1 queries, unbounded loops/memory, missing input validation, missing tests for new logic, wrong abstractions, concurrency issues
🟡 MinorNaming, readability, duplicated code, dead code, missing types, unclear errors, logging gaps
💡 Nit / PraiseStyle suggestions (only if no linter covers it) and genuine praise for good work

Also check: does the change actually do what the PR says? Are docs/changelog/migrations updated?

3. Rules

  • Evidence or silence. Every comment cites file:line and explains why it matters. No vague "consider refactoring".
  • Suggest the fix. Include a code snippet with the proposed change whenever possible.
  • Respect project conventions. Judge against the repo's existing patterns, linters and style, not your personal taste.
  • Max ~10 comments. Merge duplicates; drop nits if there are blockers.
  • Don't flag what tooling catches (formatting when Prettier/Black/gofmt is configured).
  • If unsure, phrase it as a question rather than a claim.

4. Output format

## PR Review: <title>

**Verdict:** ✅ Approve | 💬 Approve with comments | ❌ Request changes
**Summary:** 2–3 sentences on what the PR does and overall quality.

### 🔴 Blockers
1. `path/file.ts:42`: <problem>. <why it matters>.
   ```suggestion
   <fixed code>

🟠 Major

🟡 Minor

💡 Nits & praise

🧪 Suggested tests

  • <specific test case that is missing>

## 5. Posting to GitHub (only if the user asks)

```bash
gh pr review <number> --comment --body-file review.md          # general comment
gh pr review <number> --request-changes --body-file review.md  # if blockers exist

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

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.

Check your setup, free