How does the 'swarm' functionality work for code reviews?

The swarm uses multiple specialized AI agents that run in parallel. One agent might focus exclusively on security vulnerabilities while another analyzes performance impact, ensuring a more comprehensive review than a single-model approach.

Can I use this skill with GitHub Actions?

Yes, the skill is designed to be integrated into CI/CD workflows, allowing you to trigger automated reviews whenever a PR is opened, synchronized, or commented upon.

Does it require any specific local tools?

Yes, this skill requires the GitHub CLI (gh), ruv-swarm, and the Claude Flow framework to be installed and configured in your environment.

Can I customize the review agents used?

Absolutely. You can define which agents are required (e.g., security, style) and which are optional, as well as set custom thresholds for blocking PR merges based on issue severity.

Does this skill support automated code fixes?

Yes, the Style & Convention agent can be configured to auto-fix formatting, imports, and simple naming issues, providing suggested code blocks directly in the PR.

AI GitHub Code Review Swarm

Name: AI GitHub Code Review Swarm
Author: aegntic

byaegntic

•

Collaboration Tools

Orchestrates specialized AI agent swarms to perform comprehensive, multi-domain code reviews directly within GitHub pull requests.

The GitHub Code Review skill elevates your CI/CD pipeline by deploying a coordinated swarm of specialized AI agents to analyze pull requests. Unlike traditional linters, this skill utilizes domain-specific experts for security, performance, architecture, and accessibility to provide deep, contextual feedback. By integrating directly with the GitHub CLI and supporting interactive PR commands, it automates the tedious aspects of code review, enforces quality gates, and suggests intelligent fixes, allowing human reviewers to focus on high-level logic and business requirements.

Key Features

019 GitHub stars

02Automated PR management including status updates, labeling, and change requests

03Interactive PR comment commands to trigger reviews and spawn specific agents

04Multi-agent swarm coordination for parallelized, domain-specific code analysis

05Deep security auditing including OWASP, secret detection, and vulnerability analysis

06Performance profiling and architectural pattern adherence checks

Use Cases

01Automating the enforcement of architectural design patterns and organizational coding standards

02Scaling engineering teams by providing instant, expert-level feedback on every pull request

03Standardizing security and performance audits across complex microservices architectures

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add aegntic/cldcde github-code-review

For use in Claude.ai and ChatGPT

Download Skill