Can I focus the review on a specific area like security?

Yes. Upon initiation, the skill asks for your focus, allowing you to choose between a full review, security-only, performance-only, or a quick high-level review.

How many agents are used during a full review?

A full review typically launches 6 specialized agents, with an optional 7th agent specifically for LLM/AI code integration if detected.

What is the difference between Agent Teams and Task tool mode?

Agent Teams mode (Mesh) allows reviewers to message each other and cross-reference findings for deeper analysis, while Task tool mode (Star) has all reviewers report directly to a lead for faster, more cost-effective reviews.

Does this skill require the GitHub CLI?

Yes, the skill utilizes the 'gh' command-line tool to fetch PR metadata, view diffs, and check CI status automatically.

Parallel PR Reviewer

Name: Parallel PR Reviewer
Author: yonatangross

byyonatangross

•

Collaboration Tools

Conducts deep code reviews using a fleet of parallel specialized agents for quality, security, and architectural integrity.

The Parallel PR Reviewer is an advanced orchestration skill designed to simulate a senior engineering team review. By deploying 6 to 7 specialized sub-agents in parallel—including security auditors, test generators, and domain architects—it provides a multi-dimensional analysis of any pull request. The skill automates context gathering via GitHub CLI, performs deep code analysis, runs local validation checks, and synthesizes findings into a structured report. Whether you need a quick sanity check or a rigorous security audit, this skill adapts its workflow to provide production-ready feedback and catch blockers before they reach your main branch.

Key Features

01Dual orchestration modes: Mesh (Agent Teams) for cross-referencing or Star (Task tool) for efficiency

02Orchestrates 6+ specialized agents for security, quality, backend, frontend, and tests

0369 GitHub stars

04Integrated validation phase to run local linters, formatters, and test suites

05Automated context gathering using GitHub CLI for diffs, metadata, and CI status

06Comprehensive synthesis that categorizes findings into blockers and suggestions

Use Cases

01Automating the first pass of code reviews to ensure type safety and test coverage

02Standardizing PR feedback across large teams using consistent review playbooks

03Performing deep security and architectural audits on complex feature branches

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yonatangross/orchestkit review-pr

For use in Claude.ai and ChatGPT

Download Skill