Agent Code Review Swarm
Ruvnet's agent-code-review-swarm automatically identifies bugs, security flaws, and style issues in code via collaborative AI review.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add agent-code-review-swarm npx -- -y @trustedskills/agent-code-review-swarm
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"agent-code-review-swarm": {
"command": "npx",
"args": [
"-y",
"@trustedskills/agent-code-review-swarm"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
The agent-code-review-swarm skill orchestrates a swarm of AI agents to perform thorough code reviews. It leverages multiple agents, each specializing in different aspects of code quality (e.g., security, style, performance), and aggregates their findings into a consolidated report. This allows for more comprehensive feedback than a single agent could provide.
When to use it
- Large Codebase Reviews: When dealing with substantial projects where multiple perspectives are valuable.
- Security Audits: To identify potential vulnerabilities across various code segments.
- Style Guide Enforcement: Ensure consistent coding standards and best practices throughout a project.
- Pre-Merge Checks: Automate code review processes before integrating changes into the main codebase.
Key capabilities
- Multi-agent orchestration for comprehensive reviews
- Specialized agent roles (security, style, performance)
- Consolidated reporting of findings from multiple agents
Example prompts
- "Review this Python script for security vulnerabilities: [script content]"
- "Perform a code review on this JavaScript file, focusing on adherence to the Airbnb style guide: [file content]"
- "Can you analyze this Go program and identify potential performance bottlenecks? [program content]"
Tips & gotchas
The quality of the review depends heavily on the underlying agents' capabilities. Ensure the agents used are well-suited for the specific programming language and code domain being reviewed.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.