review

Orchestrates parallel subagents to inspect git diffs for logic, security, and style issues.

Updated Feb 13, 2026

Search Tags:#automation #git #quality assurance #code review #security #codex

Ready to run

npx skills add https://github.com/pegasus-isi/claude-plugin-marketplace --skill review-pegasus-isi

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: review
Source: https://github.com/pegasus-isi/claude-plugin-marketplace/tree/main/plugins/pegasus-dev/skills/review
Command: npx skills add https://github.com/pegasus-isi/claude-plugin-marketplace --skill review-pegasus-isi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses the inconsistency and oversight inherent in manual code reviews by providing a comprehensive, multi-perspective quality gate that identifies logic errors, security vulnerabilities, and stylistic issues before code is committed.

Core Features & Use Cases

Parallel Multi-Agent Analysis: Simultaneously runs simplify, security-review, and general logic checks to provide unbiased, diverse feedback.
Codex Integration: Leverages specialized Codex models for both standard and adversarial reviews to catch edge cases and potential exploits.
Autonomous Remediation: Supports an optional auto-mode for mechanical fixes, allowing for efficient cleanup of trivial issues in orchestrated pipelines.

Quick Start

Invoke the review skill to perform a comprehensive quality check on your current branch against the main branch.

Frequently Asked Questions about review

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

FAQPage Schema

How do I automate code review for security and logic defects in git branches?▼

Automated code review orchestrates parallel subagents to inspect git diffs for logic, security, and stylistic defects against a base branch. It provides consolidated, actionable feedback by running specialized checks simultaneously.

What is the best way to run multi-agent code quality checks before committing?▼

Multi-agent code quality checks run simplify, security-review, and logic checks simultaneously on your branch. This parallel approach provides unbiased, diverse feedback to identify errors, vulnerabilities, and stylistic issues before code is committed.

Can I use Codex for adversarial code review to catch edge cases?▼

Yes, Codex models support both standard and adversarial code review to catch edge cases and potential exploits. The integration leverages specialized analysis to provide comprehensive validation against identified vulnerabilities.

Does automated code review support auto-mode for fixing stylistic issues?▼

Automated code review supports an optional auto-mode for mechanical fixes. This allows efficient cleanup of trivial stylistic issues within orchestrated pipelines, streamlining the version control workflow before final commits.

Do I need a specific base branch to perform a comprehensive code quality analysis?▼

Comprehensive code quality analysis requires a specified base branch to compare diffs against. The tool operates on git-based version control workflows to validate code changes accurately.

Why does automated code review run parallel subagents instead of a single scan?▼

Automated code review runs parallel subagents to provide multi-perspective, unbiased feedback. This simultaneous execution of logic, security, and stylistic checks addresses the inconsistency and oversight inherent in manual single-scan reviews.