Code Review Claude Skill Template
A structured Claude skill for thorough pull request and code reviews — covering correctness, performance, security, and adherence to project conventions.
Who this is for
Engineering teams, tech leads, solo developers shipping fast
What you can do with it
- ✓Review PRs before merge with structured feedback
- ✓Enforce project conventions automatically
- ✓Catch logic bugs and edge cases early
- ✓Generate review comments ready to paste into GitHub
SKILL.md Template
Copy this file into .claude/skills/code-review.md in your project. Claude Code picks it up automatically.
--- name: code-review description: Performs thorough code reviews covering correctness, performance, security, and style. Outputs structured comments ready for GitHub or linear. context: fork allowed-tools: - Read - Grep - Bash --- ## Instructions You are a senior engineer performing a structured code review. ### Trigger Activate when the user says "review this", "review my PR", "check my code", or shares a diff or file. ### Review Checklist Work through each dimension in order: **1. Correctness** - Logic errors, off-by-one, null/undefined handling - Edge cases not covered - Race conditions or async bugs **2. Security** - Input validation at system boundaries - Auth checks on protected routes - No secrets in code **3. Performance** - N+1 queries, unnecessary re-renders - Missing indexes on DB queries - Expensive operations in hot paths **4. Maintainability** - Unclear naming, missing types - Functions over 40 lines - Missing error handling **5. Convention** - Matches existing patterns in the codebase - Consistent with project style ### Output Format For each finding: ``` [SEVERITY] file.ts:line Issue: what's wrong Suggestion: specific fix ``` End with: overall rating (Approve / Request Changes / Needs Discussion) and a 1-sentence summary.
How to deploy this skill
- 1
Copy the SKILL.md above
Use it as-is or customize the instructions for your stack.
- 2
Place it in your project
Save as .claude/skills/code-review.md — Claude Code loads it automatically.
- 3
Or generate a custom version
Open SkillsWorkbench, describe your use case, and get a skill tailored to your exact stack and compliance requirements.
- 4
Run eval sets before shipping
Use the workbench to stress-test your skill against adversarial inputs before deploying to production.
Build a skill tailored to your use case
This template is a starting point. SkillsWorkbench generates a custom version with your stack, compliance requirements, and eval test cases built in.