Which developer personas are included in the review?

The skill includes personas based on industry leaders: Grug (simplicity), Carmack (shippability), Ousterhout (interfaces), Beck (TDD), and Fowler (code smells).

Does it provide a visual summary of findings?

Yes, after synthesizing the review, the skill generates a self-contained HTML report and provides a command to open it directly in your browser.

Is the security check mandatory?

Yes, the 'security-sentinel' agent is mandatory for every review regardless of PR size to ensure that injection, authentication issues, and secret leaks are identified.

How does the Review Branch skill work?

It acts as a technical lead, delegating code review tasks to 12 specialized AI personas and domain specialists, then synthesizing their feedback into a single prioritized action plan.

Multi-Persona Branch Reviewer

Name: Multi-Persona Branch Reviewer
Author: phrazzld

byphrazzld

•

Collaboration Tools

Orchestrates a comprehensive code review using 12 specialized AI agents to ensure security, performance, and architectural integrity.

Review Branch transforms Claude into a technical lead that manages a rigorous code review process across a diverse team of 12 specialized AI reviewers. It leverages legendary developer personas like Carmack, Beck, and Fowler alongside mandatory security audits and performance specialists to analyze branch diffs in parallel. The skill synthesizes conflicting feedback into a prioritized, actionable plan—categorizing findings by severity—and generates a visual HTML summary report, making it an essential tool for high-stakes pre-merge quality checks.

Key Features

01Mandatory security-sentinel audit for every pull request to catch vulnerabilities

02Parallel execution of 12 specialized AI agents for rapid comprehensive analysis

03Automated visual HTML summary generation for easy review sharing

04Multi-persona feedback from legendary developers like Carmack, Beck, and Fowler

05Synthesis of conflicting viewpoints into a prioritized, deduplicated action plan

065 GitHub stars

Use Cases

01Gaining diverse architectural perspectives on complex refactoring projects

02Automating security and performance audits within the development workflow

03Orchestrating deep pre-merge quality assurance for mission-critical code

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add phrazzld/claude-config review-branch

For use in Claude.ai and ChatGPT

Download Skill