
Code Review Specialist
Reviews code for security, performance, quality and maintainability with a structured report.
- Repository
- luongnv89/claude-howto
- Stars
- 46,401
- Licence
- MIT
- Files
- 5 files
What it does
Once installed, the agent reviews code more systematically. It checks for security issues such as injection and data exposure, performance problems such as inefficient algorithms and query issues, and quality and maintainability concerns such as function length, complexity and naming. It follows a bundled checklist and writes findings in a fixed format with severity, location and suggested fixes. Two small Python helpers can compute rough metrics and compare complexity between two file versions.
Great for
- Developers who want a second set of eyes on a file or pull request before merging
- Teams that want review feedback in a consistent format with severity levels
- Anyone refactoring code who wants a rough before-and-after complexity comparison
Skip it if
- Replacing a human security audit: the checks are heuristic and the metric scripts use simple regex counts that can miss or misreport issues
- Non-Python codebases for the metric scripts: both scripts only analyze Python-style syntax, so results for other languages are unreliable
Try asking
- “Review this pull request diff for security and performance problems”
- “Analyze the code quality of utils/parser.py and rate it from 1 to 5”
- “Compare the complexity of old_module.py and new_module.py after my refactor”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 5 original files, the MIT licence and an attribution note.
code-review-specialist.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip code-review-specialist.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip code-review-specialist.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Give the agent the exact file paths or diff so it can run the scripts on the right versions
- Treat the metric scores as rough signals and verify findings yourself before acting on them
- Ask for the review checklist to be applied to one category at a time for long files
What's inside
- scripts/analyze-metrics.py
- scripts/compare-complexity.py
- SKILL.md
- templates/finding-template.md
- templates/review-checklist.md
Redistributed unmodified from luongnv89/claude-howto under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.