github-code-review

Orchestrates multi-agent code reviews on GitHub pull requests using swarm coordination.

11|Updated Jan 20, 2026
One-click install
npx skills add https://github.com/ishandutta2007/claude-agent-orchestration --skill github-code-review-ishandutta2007
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: github-code-review
Source: https://github.com/ishandutta2007/claude-agent-orchestration/tree/main/.claude/skills/github-code-review
Command: npx skills add https://github.com/ishandutta2007/claude-agent-orchestration --skill github-code-review-ishandutta2007

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manual code reviews are slow, inconsistent, and often miss security, performance, or architecture issues. This Skill automates comprehensive pull request reviews by deploying specialized review agents that analyze code in parallel and post structured feedback directly to GitHub. ## Core Features & Use Cases - Specialized Review Agents: Deploy security, performance, architecture, style, and accessibility agents that each analyze the PR diff from their domain perspective. - PR-Based Swarm Management: Create review swarms from PR data, auto-assign agents based on labels, and select topology (ring, mesh, hierarchical) based on PR size. - Quality Gates & Automation: Enforce merge-blocking thresholds, integrate with GitHub Actions CI/CD, auto-fix lint and formatting issues, and post inline review comments via the GitHub CLI. - Use Case: When a developer opens a PR touching authentication code, the skill automatically spawns security and architecture agents, runs OWASP and secrets checks, requests changes on critical findings, and labels the PR for security review. ## Quick Start Ask the assistant to review pull request 123 with security, performance, and style agents and post the findings as PR comments.

Frequently Asked Questions about github-code-review

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

FAQPage Schema
How do I run an automated code review on a GitHub pull request?

Fetch the PR data and diff with the GitHub CLI, then run npx ruv-swarm github review-init with the PR number and a list of agents such as security, performance, and style. Results are posted back to the PR as comments or formal reviews.

What review agents are available for pull request analysis?

The skill provides security, performance, architecture, style, and accessibility agents. Security checks cover OWASP issues, secrets, and CVEs; performance covers complexity and query efficiency; architecture checks SOLID principles and coupling.

Can automated reviews block a pull request from merging?

Yes. Quality gates can be defined with thresholds such as no critical security issues or less than 5% performance regression. These integrate with GitHub branch protection as required status checks like review-swarm/security.

Does this work with GitHub Actions CI/CD pipelines?

Yes. A workflow can trigger the swarm review on pull_request events, run the review with the GitHub CLI, and automatically approve, request changes, or update labels based on the review output.

Why are review agents not spawning for my pull request?

Check swarm status with npx ruv-swarm swarm-status and verify GitHub CLI authentication with gh auth status. If the swarm is stale, re-initialize it with the --force flag on the review-init command.

What are the limitations of swarm-based code review?

Large PRs can make reviews slow, so incremental mode or fewer agents may be needed. The approach also depends on external tools (GitHub CLI, ruv-swarm, claude-flow) being installed and authenticated, and token permissions must be scoped correctly.