How does this skill identify which code to review?

It automatically extracts the relevant Git SHAs by comparing your current HEAD against a specified base, such as origin/main, to analyze exactly what has changed.

Can I use this for branches other than main?

Yes, you can specify any base branch, such as 'develop' or 'staging', or even ask to review a specific number of previous commits.

What happens if the AI reviewer makes a mistake?

The workflow is designed for collaboration; you can push back with technical reasoning, provide code samples, or run tests to prove the implementation is correct.

When is the best time to trigger a code review?

It is mandatory after completing major features or before a merge, but highly recommended after every task in a subagent-driven workflow to catch issues early.

Automated Code Review

Name: Automated Code Review
Author: carlopezzuto

bycarlopezzuto

0•

Collaboration Tools

Dispatches a specialized code-reviewer subagent to validate implementations against requirements and catch issues early in the development lifecycle.

The Requesting Code Review skill automates the critical feedback loop between feature implementation and merging by deploying a specialized 'code-reviewer' subagent. It intelligently leverages Git SHAs to analyze changes, compares work against specific project requirements or documentation, and provides structured feedback on strengths and potential issues. This skill is essential for developers using subagent-driven workflows, ensuring that every task is verified and architectural integrity is maintained before technical debt or errors can cascade into later stages of the project.

Key Features

01Structured feedback categorization including Critical, Important, and Minor issues

020 GitHub stars

03Automated Git SHA extraction for precise diff analysis

04Validation against specific project plans and requirements

05Deep integration with iterative subagent-driven development workflows

06Seamless dispatch of specialized 'code-reviewer' subagents

Use Cases

01Verifying task completion in subagent-driven development before proceeding to the next step

02Gaining a fresh perspective on complex bug fixes or large-scale refactoring efforts

03Conducting a final quality check before merging feature branches into the main codebase

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add carlopezzuto/agents requesting-code-review

For use in Claude.ai and ChatGPT

Download Skill