Review code changes for correctness, security, tests, and architecture; use for code review or PR review requests (trigger keywords: code review, review changes, PR review, security review).
Scanned 9/27/2026
npx -y skills add David-Li0406/meta-skill-evloving --skill code-review-112 --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Code Review 112?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/david-li0406-code-review-112)More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.
---
name: code-review
description: "Review code changes for correctness, security, tests, and architecture; use for code review or PR review requests (trigger keywords: code review, review changes, PR review, security review)."
---
# Code Review
## What this skill does
- Reviews diffs or files for bugs, security risks, maintainability, and architecture alignment.
- Prioritizes high-severity issues and missing tests over style nitpicks.
- Adapts to repository patterns and local guidance (e.g., CLAUDE.md) when present.
## When to use it
- Trigger phrases: "code review", "PR review", "review changes", "security review".
- Use after writing or modifying code, before merge, or when asked to approve/request changes.
## Step-by-step workflow
1) Identify scope: diff, files, commit, or PR; confirm intent and expected behavior.
2) Review changed code first, then open surrounding context only when needed.
3) Scan for critical risks: correctness, security, data loss, auth/authz, injection, secrets, crypto misuse.
4) Check architecture alignment and dependency boundaries; note any layering violations.
5) Evaluate error handling, logging, resource cleanup, and edge cases.
6) Assess code quality and maintainability: clarity, naming, abstraction level, duplication, complexity, and dependency changes; note SOLID issues only when material.
7) Check tests and coverage gaps; call out missing or brittle tests.
8) Flag performance issues only when evidence exists; avoid speculative optimizations.
9) Write actionable feedback with file references and minimal fix guidance.
10) Summarize remaining risks and verification steps.
## Expected outputs / formatting
- Findings ordered by severity with file paths.
- Short rationale and concrete fix suggestions for each issue.
- Call out security items with optional CWE/OWASP references when applicable.
- Missing tests called out explicitly.
- Brief change summary only after findings.
- Keep feedback pragmatic; avoid theoretical or stylistic nitpicks.
## Example prompts
- "Do a code review of the staged changes."
- "Review this PR for security and correctness issues."
- "Check these files for architecture violations."
- "Review the code you just generated and call out risks."
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!