review-engine

Automate code reviews by analyzing changes with security and performance criteria.

1|Updated Jun 3, 2026
One-click install
npx skills add https://github.com/SchSeba/ai-plugins --skill review-engine-schseba
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-engine
Source: https://github.com/SchSeba/ai-plugins/tree/main/skills/review-engine
Command: npx skills add https://github.com/SchSeba/ai-plugins --skill review-engine-schseba

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires git, codegraph, Subagent, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the complexity and time-consuming nature of code reviews by providing a comprehensive, multi-perspective review engine.

Core Features & Use Cases

  • Multi-Perspective Reviews: Offers security, performance, language-specific, and other relevant reviews.
  • Parallel Reviewers: Utilizes specialized sub-agents for parallel reviews to reduce time and ensure thoroughness.
  • Cross-Cutting Analysis: Performs end-to-end user flow tracing, state assumption auditing, and data boundary following.
  • Verdict Rendering: Provides a structured summary of findings and a clear verdict on the code changes.
  • Use Case: For a developer working on a large codebase, this Skill can be integrated into their workflow to ensure code quality and security.

Quick Start

Run the review-engine skill with the command: /review-engine review https://github.com/owner/repo/pull/42

Frequently Asked Questions about review-engine

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

FAQPage Schema
How do I automate code reviews for pull requests?

Automating code reviews for pull requests involves analyzing code changes, applying security and performance criteria, and orchestrating parallel reviews. This engine uses sub-agents to provide a structured summary and a clear verdict on the changes.

What is multi-perspective code review and how does it work?

Multi-perspective code review analyzes code changes by applying security, performance, and language-specific criteria simultaneously. It utilizes specialized sub-agents for parallel reviews to reduce time and ensure thoroughness across different analysis dimensions.

Does the parallel code review engine work with local Git diffs?

Yes, the parallel code review engine works with local Git diffs. It requires Git and various code analysis tools to process changes, performing cross-cutting analysis like end-to-end user flow tracing and state assumption auditing.

Can I use sub-agents to parallelize security and performance reviews?

Yes, you can use sub-agents to parallelize security and performance reviews. The engine utilizes specialized sub-agents to conduct parallel reviews, reducing time and ensuring thoroughness across multiple analysis perspectives.

What's the best way to trace user flows and audit state assumptions during code reviews?

The best way to trace user flows and audit state assumptions is through cross-cutting analysis. The engine performs end-to-end user flow tracing, state assumption auditing, and data boundary following to deliver a structured summary of findings.

Do I need Git and code analysis tools to run automated code reviews?

Yes, you need Git and various code analysis tools to run automated code reviews. These dependencies are required to gather context and orchestrate the parallel reviews for security, performance, and language-specific criteria.