review

Automate two-stage AI review of code changes for defects and security gaps.

6|Updated Apr 20, 2026
One-click install
npx skills add https://github.com/waitdeadai/minmaxing --skill review-waitdeadai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/waitdeadai/minmaxing/tree/main/.claude/skills/review
Command: npx skills add https://github.com/waitdeadai/minmaxing --skill review-waitdeadai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Two-stage AI review catches defects, security issues, and quality problems early, then human sign-off validates business logic, UX, and edge cases.

Core Features & Use Cases

  • AI-driven analysis identifies logic errors, security risks, and style violations.
  • Human sign-off ensures alignment with product requirements and UX considerations.
  • Swarm-style review supports parallel lanes up to MAX_PARALLEL_AGENTS for large changes.

Quick Start

Run a review on a PR by issuing a swarm review request to /review with the changed files.

Frequently Asked Questions about review

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

FAQPage Schema
How does two-stage AI code review work for pull requests?

Two-stage AI code review first uses automated analysis to identify defects, security gaps, and quality issues in pull requests, then requires human sign-off to validate business logic, UX, and edge cases before release.

Can I run AI code review on large diffs using parallel agents?

Yes, AI code review supports swarm-style analysis with parallel lanes up to MAX_PARALLEL_AGENTS, allowing large diffs and file groups to be processed simultaneously before human sign-off.

How do I automate security and quality checks on changed files?

You can automate security and quality checks by issuing a swarm review request with your changed files, triggering AI-driven analysis to detect logic errors, security risks, and style violations.

What is the difference between automated analysis and human sign-off in code review?

Automated analysis identifies code-level defects, security gaps, and style violations, while human sign-off validates product requirement alignment, UX considerations, and complex edge cases for release-ready changes.

Does AI code review support staged reviews for escalating code changes?

Yes, AI code review defines a two-stage workflow with staged reviews and clear criteria for escalation, ensuring complex changes receive appropriate automated analysis and human validation.