Where are the review results stored?

Reviews are saved as detailed markdown documents within the .claude/plans/ directory of your repository for easy reference and tracking.

What specific issues does this code review skill look for?

It performs a comprehensive check for thread safety, async safety, security vulnerabilities, data integrity, test quality, and general code smells like hard-coded configurations.

Can I customize the review criteria?

Yes, the skill follows a customizable template located at prompts/CODE_REVIEW.md, allowing you to tailor the audit standards to your specific project needs.

How does it identify which code needs to be reviewed?

The skill uses git status, git diff, and git log to automatically determine uncommitted changes or commits made since the last recorded review point.

Does it provide actual code fixes or just descriptions?

Unlike basic linters, this skill provides concrete 'before and after' code fixes with specific file and line references to streamline the remediation process.

Professional Code Reviewer

Name: Professional Code Reviewer
Author: MQFacultyOfArts

byMQFacultyOfArts

Collaboration Tools

Conducts rigorous, high-standard code reviews focusing on security, concurrency, and architectural integrity.

About

The Professional Code Reviewer skill transforms Claude into a pedantic and expert auditor for your codebase. It systematically analyzes recent changes—from uncommitted diffs to recent logs—to identify critical vulnerabilities, race conditions, and technical debt. By adhering to a 'no quick hacks' policy and providing line-specific fixes, this skill ensures your code meets production-grade standards for thread safety, async patterns, and comprehensive test coverage before any significant commit or merge.

Key Features

Detection of code smells, magic numbers, and hard-coded configurations
Automated audit of thread and async safety to prevent complex race conditions
Security vulnerability scanning and data integrity verification for recent changes
Enforcement of a 'No Quick Hacks' policy for long-term architectural health
0 GitHub stars
Detailed review reports with file:line references and concrete code fixes

Use Cases

Pre-commit audits for significant feature additions or structural refactors
Identifying concurrency bugs in complex asynchronous applications
Generating structured feedback and fix suggestions during the final phase of a development sprint

About

Key Features

Detection of code smells, magic numbers, and hard-coded configurations
Automated audit of thread and async safety to prevent complex race conditions
Security vulnerability scanning and data integrity verification for recent changes
Enforcement of a 'No Quick Hacks' policy for long-term architectural health
0 GitHub stars
Detailed review reports with file:line references and concrete code fixes

Use Cases

Pre-commit audits for significant feature additions or structural refactors
Identifying concurrency bugs in complex asynchronous applications
Generating structured feedback and fix suggestions during the final phase of a development sprint