What is the primary benefit of the Requesting Code Review skill?

It automates the verification process, ensuring that your implementation matches your initial plan and catching critical issues early in the development cycle.

Should I use this for small tasks or only large features?

While mandatory for major features, it is highly recommended for every task in a subagent-driven workflow to catch errors before they compound.

How does the subagent identify what to review?

It uses Git SHAs (Base and Head) to target specific changes and references a provided plan or requirements document to ensure all criteria are met.

What happens if the AI reviewer makes a mistake?

The workflow encourages developers to push back with technical reasoning or test evidence if the reviewer's feedback is incorrect, ensuring a collaborative and accurate result.

Does this skill require a specific git configuration?

It requires a standard git repository where you can identify a Base SHA (starting point) and a Head SHA (current state) to define the review scope.

Automated Code Review Agent

Name: Automated Code Review Agent
Author: seanGSISG

byseanGSISG

0•

협업 도구

Dispatches a specialized code-reviewer subagent to audit implementations against requirements and git history before merging.

The Requesting Code Review skill streamlines the development lifecycle by enabling Claude to perform rigorous self-reviews or peer-style audits via a dedicated reviewer subagent. By comparing current implementations against a baseline using Git SHAs, it identifies logic errors, missing requirements, and architectural inconsistencies before code is finalized. This skill is essential for developers using subagent-driven workflows who want to maintain high code quality and prevent technical debt from cascading through complex feature builds.

주요 기능

01Verification of implementation against specific project requirements

02Seamless integration with task-based development plans

03Git SHA-based differential analysis between commits

04Automated subagent dispatch for deep code auditing

05Categorized feedback loops for Critical, Important, and Minor issues

060 GitHub stars

사용 사례

01Gaining an objective perspective on complex bug fixes or refactoring projects

02Conducting a final quality audit before merging feature branches into main

03Verifying individual tasks in subagent-driven development to prevent cascading errors

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add seangsisg/crispy-claude requesting-code-review

For use in Claude.ai and ChatGPT

Download Skill