What are the red flags to avoid when using this skill?

Never skip reviews for 'simple' tasks, ignore critical issues, or proceed with unfixed important issues. Consistent review is key to preventing technical debt.

How should I handle feedback from the AI code reviewer?

Fix critical and important issues immediately. Minor issues can be noted for later. If you believe the reviewer is incorrect, you should push back with technical reasoning and code evidence.

How does the subagent know what code to review?

The skill utilizes git SHAs (Base and Head) to isolate the specific commits and changes made during a task, ensuring the reviewer focuses only on the relevant code deltas.

When is the best time to request a code review with this skill?

It is mandatory after each task in subagent-driven development, after completing major features, or before merging to the main branch. It is also valuable when you are stuck or before starting a refactor.

Code Review Automator

Name: Code Review Automator
Author: cloudfieldcz

bycloudfieldcz

•

コラボレーションツール

Automates structured code reviews by dispatching subagents to analyze changes and provide actionable feedback before merging.

The Code Review Automator is a specialized Claude Code skill designed to integrate quality assurance directly into the AI-driven development workflow. By dispatching the cf-powers:code-reviewer subagent, it allows developers to catch architectural issues and bugs before they cascade. The skill uses git SHAs to isolate specific changes, providing a focused review based on implementation plans and requirements. It is an essential tool for subagent-driven development, ensuring that every task meets quality standards before proceeding to the next phase.

主な機能

01Git-based delta identification using Base and Head SHAs

022 GitHub stars

03Categorized feedback across Critical, Important, and Minor levels

04Requirement-based verification against implementation plans

05Automated subagent dispatch for deep code analysis

06Seamless integration with subagent-driven development workflows

ユースケース

01Verifying major feature implementations before merging into the main branch

02Getting a fresh perspective on complex bug fixes or large-scale refactors

03Performing incremental reviews after completing individual tasks in a development plan

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add cloudfieldcz/cf-powers requesting-code-review

For use in Claude.ai and ChatGPT

Download Skill