pr-review

Use this skill to review pull requests against coding standards and best practices. Invoke when reviewing code changes before merge.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "pr-review" with this command: npx skills add gentamura/dotfiles/gentamura-dotfiles-pr-review

PR Review

Systematic code review following team standards.

Review Process

1. Understand the Change

  • Read PR description and linked issues
  • Understand the intent and scope
  • Check if the approach aligns with architecture

2. Review Code Quality

Go through each checklist section below.

3. Provide Feedback

  • Be specific and actionable
  • Explain the "why" behind suggestions
  • Distinguish blocking issues from suggestions
  • Acknowledge good patterns

Review Checklist

TypeScript

  • No any types used
  • No as casts (type guards used instead)
  • Functions have explicit return types
  • Proper error types defined
  • No TypeScript errors (bun run build passes)

React

  • useEffect only for external synchronization
    • API calls, WebSocket, browser APIs, timers
  • No derived state in useEffect
  • No props copied to state
  • User actions handled in event handlers
  • Effects have comments explaining external resource
  • Proper cleanup in effects when needed

Code Quality

  • bun run lint:fix (or bun run lint if no fix script) passes
  • No console.log in production code
  • Follows existing patterns in codebase
  • No dead code or commented-out code
  • Meaningful variable and function names

Security

  • Input validated at boundaries
  • No hardcoded secrets or credentials
  • Proper authentication/authorization checks
  • No SQL injection vulnerabilities
  • No XSS vulnerabilities

Database (if applicable)

  • Existing migrations NOT edited
  • Schema changes via drizzle-kit generate
  • Appropriate indexes considered
  • No N+1 query patterns

Testing

  • Tests cover happy path
  • Tests cover error cases
  • Tests are meaningful (not just coverage)

Documentation

  • Comments and docs in English
  • Complex logic explained
  • API changes documented
  • README updated if needed

Feedback Template

## Summary

[Overall assessment: Approve / Request Changes / Comment]

## Blocking Issues

- [ ] **[File:Line]** Issue description
  - Why it's a problem
  - Suggested fix

## Suggestions

- **[File:Line]** Suggestion description
  - Why this would be better

## Questions

- [Question about design or implementation]

## Praise

- [Acknowledge good patterns or improvements]

Severity Levels

LevelMeaningAction
BlockingMust fix before mergeRequest changes
SuggestionWould improve codeComment
NitpickMinor style preferenceOptional
QuestionNeed clarificationComment

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

done

No summary provided by upstream source.

Repository SourceNeeds Review
General

incident-hotfix

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

agent-browser

No summary provided by upstream source.

Repository SourceNeeds Review
General

architecture-adr

No summary provided by upstream source.

Repository SourceNeeds Review