code-review
Code Review Skill
Guidelines for thorough and effective code reviews.
When to Apply
Use these guidelines when:
- Reviewing pull requests
- Refactoring existing code
- Conducting code audits
- Ensuring code quality standards
Review Categories
1. Code Quality
- Check for consistent code style
- Ensure proper error handling
- Verify meaningful variable and function names
- Look for code duplication
2. Security
- Check for SQL injection vulnerabilities
- Verify input validation and sanitization
- Review authentication and authorization logic
- Check for sensitive data exposure
3. Performance
- Identify potential performance bottlenecks
- Check for unnecessary database queries
- Review algorithm efficiency
- Look for memory leaks
4. Testing
- Verify test coverage
- Check test quality and relevance
- Ensure edge cases are covered
- Review mocking and fixtures
5. Documentation
- Check for code comments
- Verify README updates
- Review API documentation
- Ensure changelog is updated
Review Process
-
Initial Review
- Read through the changes
- Understand the purpose
- Identify major issues first
-
Detailed Review
- Check each file
- Verify logic and implementation
- Test if necessary
-
Final Review
- Summarize findings
- Provide actionable feedback
- Verify fixes if needed
Best Practices
- Be constructive and respectful
- Explain the "why" behind comments
- Suggest improvements, not just problems
- Acknowledge good work
- Keep reviews timely
More from uwe-schwarz/skills
github-pr-review-workflow
Complete workflow for handling GitHub PR reviews using gh pr-review extension
40coderabbit-review
Perform code review using CodeRabbit. Must use this skill if instructed to perform a CodeRabbit review.
9documentation
Automated documentation generation and management. Use for creating, updating, and maintaining project documentation including README files, API docs, and inline code comments.
8example-skill
An example skill demonstrating the Agent Skills format. Use this as a template for creating your own skills.
8cost-estimate
Estimate the replacement cost, engineering effort, calendar time, and market-rate value of a software project by analyzing a repository's size, architecture, complexity, and delivery maturity. Use when asked for a build-cost estimate, rebuild quote, delivery timeline, stakeholder-facing budget range, pricing sanity check, or ROI/value analysis for work produced by Claude, Codex, OpenCode, Crush, or another AI coding agent.
1