code-review

Audit recent code changes for security, thread safety, and test quality.

1|1|Updated Jan 13, 2026
One-click install
npx skills add https://github.com/MQFacultyOfArts/PromptGrimoireTool --skill code-review-mqfacultyofarts
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/MQFacultyOfArts/PromptGrimoireTool/tree/main/.claude/skills/code-review
Command: npx skills add https://github.com/MQFacultyOfArts/PromptGrimoireTool --skill code-review-mqfacultyofarts

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a disciplined, professional code-review workflow that systematically evaluates recent changes for thread safety, async safety, security vulnerabilities, data integrity, test quality, and code smells, reducing risk before commits.

Core Features & Use Cases

  • Mandatory review criteria: loads a formal review template (prompts/CODE_REVIEW.md) and applies all criteria.
  • Scope and timing: reviews commits since the last review, before committing significant changes, after spikes, or when prompted by the Stop hook.
  • Deliverables: produces a structured review document with executive summary, issues grouped by severity (with file:line references), concrete before/after fixes, a fix checklist, and verification steps saved to .claude/plans/.
  • Traceability: preserves references to the original changes and relevant files for auditing.

Quick Start

Use the code-review skill to evaluate the latest changes in your working directory and save the plan to .claude/plans/.

Frequently Asked Questions about code-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a pre-commit code review that checks thread safety and security vulnerabilities?

A pre-commit code review audits recent changes for thread safety, async safety, security vulnerabilities, and data integrity by loading formal review criteria and outputting a structured fix plan with file:line references to .claude/plans/.

What is the best way to automatically review code quality after a spike or rapid prototyping session?

The best way to review code quality after a spike is triggering a maximally thorough code review that evaluates changes for code smells and test quality, enforcing a no-quick-hacks policy and generating a verification checklist.

How does a professional code review enforce a no-quick-hacks policy before committing changes?

A professional code review enforces a no-quick-hacks policy by systematically applying mandatory review criteria from a template, auditing changes for security vulnerabilities and data integrity, and producing concrete before/after fixes.

Can I use this code review process to evaluate test quality and code smells automatically?

Yes, this code review process evaluates test quality and identifies code smells automatically by auditing recent commits, grouping issues by severity, and generating an executive summary with traceable file:line references.

What deliverables should I expect from a structured code review of recent git changes?

Deliverables from a structured code review include an executive summary, issues grouped by severity with file:line references, before/after fixes, a fix checklist, and verification steps saved as a traceable plan document.

When do I need a formal code review for thread safety and async safety issues?

You need a formal code review for thread safety and async safety issues before committing significant changes, after rapid prototyping spikes, or when prompted by a Stop hook to ensure data integrity and reduce risk.