What kind of feedback does the subagent provide?

The subagent returns a structured assessment including project strengths and prioritized issues labeled as Critical, Important, or Minor.

When should I use the requesting-code-review skill?

It is best used after completing any major task, before merging to the main branch, or whenever you need a fresh technical perspective on your implementation.

How does the reviewer identify what to check?

The skill uses git SHAs (BASE_SHA and HEAD_SHA) to pinpoint and analyze the specific changes made since the last stable commit or task start.

Can I disagree with the subagent's feedback?

Yes. If the reviewer provides incorrect technical feedback, you should push back with technical reasoning or code evidence to clarify the implementation.

Why is review mandatory in subagent-driven development?

Frequent reviews prevent 'hallucination cascades' where small errors in early tasks compound into major architectural failures in later stages.

AI Code Reviewer Subagent

Name: AI Code Reviewer Subagent
Author: SwervoJake

bySwervoJake

0•

コラボレーションツール

Automates technical code reviews by dispatching a specialized subagent to identify bugs and verify requirements before merging.

The requesting-code-review skill streamlines the quality assurance process within Claude Code by leveraging a dedicated subagent to perform technical audits. It enforces a 'review early, review often' philosophy, allowing developers to catch architectural flaws and critical bugs before they propagate through a project. By utilizing git SHAs to define the scope of review and providing structured feedback categorized by severity, this skill ensures that subagent-driven development remains high-quality and aligned with project requirements.

主な機能

01Structured templates for consistent review reporting

02Git SHA-based diff analysis for precise review scoping

03Automated subagent dispatch for technical code auditing

04Integration with subagent-driven development workflows

050 GitHub stars

06Categorized feedback tiers: Critical, Important, and Minor

ユースケース

01Performing mandatory health checks before merging code into the main branch

02Verifying work meets requirements after completing a major feature task

03Obtaining a fresh technical perspective when stuck on complex implementation details

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add swervojake/claaaude-studio requesting-code-review

For use in Claude.ai and ChatGPT

主な機能

01Structured templates for consistent review reporting

02Git SHA-based diff analysis for precise review scoping

03Automated subagent dispatch for technical code auditing

04Integration with subagent-driven development workflows

050 GitHub stars

06Categorized feedback tiers: Critical, Important, and Minor

ユースケース

01Performing mandatory health checks before merging code into the main branch

02Verifying work meets requirements after completing a major feature task

03Obtaining a fresh technical perspective when stuck on complex implementation details

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add swervojake/claaaude-studio requesting-code-review

For use in Claude.ai and ChatGPT