What do the severity levels mean?

'MUST FIX' indicates critical issues or bugs, 'SHOULD FIX' suggests industry best practices, and 'CONSIDER' offers optional improvements or structural simplifications.

Does it review already committed code?

This specific skill is optimized for reviewing uncommitted and staged changes to ensure quality before code is officially added to the repository history.

Will it provide positive feedback?

Yes, if no issues are identified during the analysis, the skill will confirm that your code looks solid and meets the checklist requirements.

Can I customize the review criteria?

Yes, the skill applies a checklist located in 'references/checklist.md', which can be tailored to include your specific project standards or architectural requirements.

How does the skill access my code changes?

The skill utilizes standard 'git diff' and 'git diff --cached' commands to examine your current uncommitted and staged work directly within your environment.

Automated Code Review

Name: Automated Code Review
Author: denisraison

bydenisraison

•

Collaboration Tools

Analyzes uncommitted Git changes against a standardized checklist to identify quality issues and architectural gaps.

The Review skill automates the process of checking code changes before they are committed, ensuring high standards and consistency across your codebase. By examining both staged and unstaged changes through Git diffs, it applies a specialized checklist to catch potential bugs, simplification opportunities, and logic gaps. Each finding is reported with specific file and line references along with a severity rating—ranging from critical 'Must Fix' issues to optional 'Consider' suggestions—making it an essential tool for maintaining code health and conducting pre-commit sanity checks.

Key Features

01Checklist-based quality assessment from local references

02Precise file and line number references for every finding

03Severity-labeled issue reporting (Must Fix, Should Fix, Consider)

04Automatic Git diff and cached diff analysis

051 GitHub stars

06Sanity checks for logic gaps and code simplification

Use Cases

01Conducting a thorough pre-commit sanity check on local changes

02Identifying simplification opportunities in complex diffs before a PR

03Ensuring new code adheres to project-specific quality standards

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add denisraison/claude-plugins review

For use in Claude.ai and ChatGPT

Download Skill