code-reviewer_skill

Overview

This skill performs comprehensive code reviews with a focus on best practices, security, and performance. It is optimized for Rust projects but supports multi-language reviews when needed. The goal is to produce clear, prioritized feedback that teams can act on quickly.

How this skill works

I start with a high-level assessment of structure, design patterns, and naming to spot architectural or maintainability issues. Then I run a detailed analysis covering security (common vulnerabilities), performance hotspots, style and conventions, and documentation and tests. Findings are grouped by severity and include concrete, actionable recommendations and code examples when helpful. Each review highlights positives as well as remediation steps prioritized by impact.

When to use it

• Before merging a feature or refactor to catch regressions and security flaws early
• During security audits to identify common vulnerabilities in application logic
• When optimizing critical code paths for latency or memory usage
• When onboarding new contributors to ensure codebase consistency
• To evaluate test coverage and suggest missing or brittle tests

Best practices

• Prioritize fixes by severity: address Critical and High issues first
• Provide short code examples or diffs for remediation to reduce friction
• Balance ideal solutions with pragmatic trade-offs suitable for the project
• Be constructive: highlight good patterns and explain reasoning for changes
• Reference relevant Rust idioms and crates when recommending improvements

Example use cases

• Reviewing a Rust pull request that introduces unsafe code and concurrency changes
• Evaluating performance regressions after a dependency upgrade
• Auditing input handling to prevent injection or parsing vulnerabilities
• Assessing test gaps for newly added modules and suggesting unit/integration tests
• Standardizing naming, error handling, and module layout across a codebase