code-review

Coordinate multi-agent code review for GitHub pull requests with severity-based findings.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/kulharir7/karya --skill code-review-kulharir7
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/kulharir7/karya/tree/main/workspace/plugins/code-review
Command: npx skills add https://github.com/kulharir7/karya --skill code-review-kulharir7

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automate and accelerate code reviews by orchestrating multiple specialized agents to review GitHub PRs in parallel.

Core Features & Use Cases

  • Spawns parallel agents (Security, Logic, Performance, Style) to inspect changes
  • Aggregates findings, ranks by severity, and generates a comprehensive report
  • Optional: auto-comment on PRs and provide inline feedback to developers

Quick Start

Run /code-review owner/repo#PR to start an automated, multi-agent PR review.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I automate GitHub pull request reviews?

You can automate GitHub pull request reviews by orchestrating multiple parallel agents to inspect diffs, aggregate findings by severity, and generate a comprehensive report with optional inline comments.

How does multi-agent code review work for pull requests?

Multi-agent code review works by spawning parallel agents focused on security, logic, performance, and style to inspect PR changes, adapting review depth based on PR size and metadata.

Can I use automated code review for large GitHub PRs?

Yes, automated code review supports large GitHub PRs by evaluating changed files and metadata to adapt review depth, ensuring comprehensive analysis across security, logic, performance, and style.

What's the best way to find security and performance issues in a PR?

The best way to find security and performance issues in a PR is to deploy specialized parallel agents that evaluate diffs and surface actionable findings ranked by severity in a structured report.

Do I need to install dependencies to run multi-agent PR review?

No dependencies are required to run multi-agent PR review. You can start an automated review directly by running the command with the owner, repo, and PR number.

What are the limitations of automated code review agents?

Automated code review agents focus on evaluating diffs and metadata for security, logic, performance, and style, but do not execute the code or test runtime behavior, limiting findings to static analysis.