code-review

Review code changes with severity-rated findings and merge gating rules.

3|Updated May 24, 2026
One-click install
npx skills add https://github.com/shayne-snap/oh-my-antigravity --skill code-review-shayne-snap
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/shayne-snap/oh-my-antigravity/tree/main/plugin/skills/code-review
Command: npx skills add https://github.com/shayne-snap/oh-my-antigravity --skill code-review-shayne-snap

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Code-review turns proposed changes into clear, evidence-backed feedback so you can catch security and quality issues before they reach production or main branches.

Core Features & Use Cases

  • Comprehensive, severity-rated review: Produces CRITICAL/HIGH/MEDIUM/LOW findings with concrete file:line evidence.
  • Two-lane evaluation for correctness and design: Runs a parallel code-reviewer lane for implementation and a architect lane for tradeoffs, with an optional security-reviewer lane for security-sensitive scopes.
  • Deterministic merge gating rules: Converts review outcomes into REQUEST CHANGES or COMMENT using explicit blocker contracts, not vibes.

Quick Start

Run a read-only review of your current change set by asking: /omagy:code-review

Frequently Asked Questions about code-review

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

FAQPage Schema
What is severity-rated code review and how does it enforce merge gating?

Severity-rated code review assigns CRITICAL, HIGH, MEDIUM, or LOW findings to code changes with concrete file:line evidence. It enforces deterministic merge gating by converting evidence-backed CLEAR, WATCH, or BLOCK outcomes into strict REQUEST CHANGES or COMMENT actions.

How do I run a comprehensive code review on a git diff or pull request?

To run a comprehensive code review on a git diff or pull request, execute a read-only review command against your current change set. The process evaluates implementation correctness and architectural tradeoffs across parallel review lanes to produce merge-ready feedback.

Can I use automated code review for broad project audits instead of just PR diffs?

Yes, automated code review can be applied to broad project audits in addition to PR-scoped diff reviews. The evaluation enforces strict in-scope inspection boundaries to ensure feedback remains relevant to the targeted changes or overall project architecture.

Does automated code review evaluate architectural tradeoffs or just implementation quality?

Automated code review evaluates both implementation quality and architectural tradeoffs. It uses parallel evaluation lanes, deploying a code-reviewer for implementation details and an architect for design tradeoffs, with an optional security-reviewer for security-sensitive scopes.

What is the best way to catch security and maintainability issues before merging to main?

The best way to catch security and maintainability issues before merging is to use a multi-lane code review process. It identifies vulnerabilities and maintainability flaws with concrete evidence, applying deterministic gating rules to block problematic merges.