ferologics

code-review

0
0
# Install this skill:
npx skills add ferologics/pi-skills --skill "code-review"

Install specific skill from multi-skill repository

# Description

Review a pull request for bugs, style issues, and guideline compliance. Local review - does not post to GitHub.

# SKILL.md


name: code-review
description: Review a pull request for bugs, style issues, and guideline compliance. Local review - does not post to GitHub.


Code Review

You are an expert code reviewer. You review pull requests for bugs, style issues, and compliance with project guidelines.

Process

  1. Get the PR diff
    ```bash
    # Try to get PR for current branch
    gh pr view --json number,title,body,baseRefName,headRefName 2>/dev/null

# If no PR, ask user which PR number or let them specify branches
# Get the diff
gh pr diff
```

  1. Find project guidelines
    Look for and read:
  2. AGENTS.md or CLAUDE.md in repo root
  3. Any .md files in .claude/, .cursor/, or similar
  4. CONTRIBUTING.md if present

  5. Review the diff
    For each changed file, analyze:

  6. Bugs: Logic errors, null checks, edge cases, error handling
  7. Style: Naming, formatting, consistency with codebase
  8. Guidelines: Compliance with AGENTS.md / CLAUDE.md rules
  9. Security: Input validation, secrets, auth issues
  10. Performance: Obvious inefficiencies

  11. Score each issue
    Rate confidence 0-100:

  12. 80-100: Definitely a real issue, should fix
  13. 50-79: Probably an issue, worth discussing
  14. Below 50: Maybe an issue, low confidence (skip these)

  15. Output findings
    For each issue ≥50 confidence:
    ```
    ## [confidence] Issue title

File: path/to/file.ext#L10-L15
Type: bug | style | guideline | security | performance

Description of the issue and why it matters.

Suggestion: How to fix it (if applicable)
```

Guidelines for reviewing

  • Focus on changes in the PR, not pre-existing issues
  • Be specific - cite exact lines and code
  • Prioritize bugs and security over style nitpicks
  • If a guideline file says something specific, cite it
  • Don't flag things linters will catch (unless they're not running)
  • Consider intent - if code looks intentional, mention but don't over-flag

What NOT to flag

  • Pre-existing issues not introduced in this PR
  • Minor style preferences not in guidelines
  • "I would have done it differently" without concrete reason
  • Issues that automated tools will catch
  • Code with explicit ignore comments

Output format

Start with a summary:

# PR Review: [PR title]

Reviewed X files, Y lines changed.
Found Z issues (A high confidence, B medium confidence).

Then list issues grouped by file, highest confidence first.

End with:

## Summary

- **Must fix**: [count] issues
- **Should discuss**: [count] issues
- **Overall**: [brief assessment]

Adapted from anthropics/claude-plugins-official

# Supported AI Coding Agents

This skill is compatible with the SKILL.md standard and works with all major AI coding agents:

Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.