review-team

Orchestrate multi-step code reviews with expert team and GitHub API.

54|16|Updated Mar 10, 2026
One-click install
npx skills add https://github.com/EricTechPro/startup-claude-skills --skill review-team-erictechpro
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-team
Source: https://github.com/EricTechPro/startup-claude-skills/tree/main/skills/review-team
Command: npx skills add https://github.com/EricTechPro/startup-claude-skills --skill review-team-erictechpro

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires gh, claude, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the process of code review in software development by organizing a team of experts and introducing an adversarial challenge phase, ensuring thorough and reliable feedback.

Core Features & Use Cases

  • Expert Review Team: Assembles a team of 5 specialists for a comprehensive code review.
  • Adversarial Challenge: A Devil's Advocate scrutinizes each finding to filter false positives and improve accuracy.
  • Use Case: When you need to ensure that critical PRs receive a high-quality review, this skill can be instrumental in providing a confident, adversarially filtered report.

Quick Start

/review-team https://github.com/owner/repo/pull/123

Frequently Asked Questions about review-team

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate GitHub PR review with an expert team and adversarial challenge?

Automated GitHub PR review uses a multi-step process orchestrating a five-member expert team and a Devil's Advocate to scrutinize findings. It utilizes the GitHub API for PR metadata and integrates with Claude Code's Task and SendMessage tools for parallel review, defense, and final report synthesis.

What is adversarial code review and how does a Devil's Advocate filter false positives?

Adversarial code review introduces a Devil's Advocate phase that actively challenges each finding from the expert team. This defense and rebuttal mechanism filters out false positives and improves the accuracy of the final synthesized PR review report.

Do I need GitHub CLI and Claude Code to run multi-step code review workflows?

Yes, running this multi-step code review workflow requires both the GitHub CLI (gh) and Claude Code (claude) dependencies. The skill uses these tools to fetch PR metadata, manage parallel review communication, and synthesize the final adversarially filtered report.

How do I configure project-specific code review for GitHub pull requests with Jira integration?

Configuring project-specific PR review with Jira integration requires setting up dedicated project configuration files. This setup ensures the expert team review process can successfully fetch relevant ticket context and PR metadata to accurately synthesize the final report.

What is the best way to ensure critical PRs receive a high-quality code review?

The best way to ensure critical PRs receive high-quality review is assembling a specialist expert team followed by an adversarial challenge phase. This approach guarantees a confident, adversarially filtered report by combining parallel review with a defense and rebuttal mechanism.

Why does my automated PR review workflow require project-specific configuration?

Automated PR review workflows require project-specific configuration to accurately integrate Jira for ticket context and properly route expert team communication. Without this setup, the multi-step orchestration cannot fetch the necessary PR metadata or synthesize a reliable final report.