What problem does it solve?
Manual code reviews can be time-consuming, inconsistent, and prone to human error. This skill automates the initial stages of code review, providing immediate, structured feedback on correctness, style, testing, and security, allowing human reviewers to focus on higher-level architectural concerns.
Core Features & Use Cases
- Automated Pre-Review Checks: Automatically runs linters (
just lint-for-claude) and tests (just test-for-claude) to catch common issues before human review.
- Comprehensive Review Checklist: Evaluates code against a detailed checklist covering correctness, quality, testing, performance, security, and maintainability.
- Structured Feedback: Generates a clear, actionable review report with critical issues, suggestions, positive notes, and action items, including file and line references.
- Use Case: Before submitting a pull request, use
/review-code. The skill will run all checks, analyze your changes, and provide a detailed report, helping you catch and fix issues proactively, leading to faster approvals and higher quality code.
Quick Start
Ask Claude to "Review my staged code changes and provide feedback."