review

Perform two-pass code reviews of git diffs to detect structural bugs and edge cases.

204|14|Updated Mar 25, 2026
One-click install
npx skills add https://github.com/garagon/nanostack --skill review-garagon
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/garagon/nanostack/tree/main/review
Command: npx skills add https://github.com/garagon/nanostack --skill review-garagon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill addresses the risk of merging changes that introduce structural bugs, security issues, or scope drift by applying a disciplined two-pass review: a structural correctness pass followed by an adversarial edge-case pass that scales with diff size.

Core Features & Use Cases

  • Two-pass methodology: First verify correctness and consistency, then adopt an adversarial mindset to probe for race conditions, injection risks, and failure modes.
  • Mode-scaled intensity: Supports quick, standard, and thorough modes with auto-suggestion based on diff size and sensitive paths.
  • Integrations & artifacts: Reads plan artifacts, detects scope drift, runs security suggestion hooks, and saves a review artifact for downstream autopilot steps.
  • Use case: Run this on a feature PR touching auth or payments to catch design issues, regressions, and security surface changes before shipping.

Quick Start

Run /review in your agent to perform a two-pass code review on the current changes and produce a findings artifact.

Frequently Asked Questions about review

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

FAQPage Schema
How do I perform an automated code review on a git diff to catch security issues?

Automated code review on git diffs applies a two-pass methodology to detect structural bugs and adversarial edge cases, including security-sensitive injection risks. It analyzes the changes for correctness and consistency before probing for potential failure modes.

What is the best way to review code changes for race conditions before merging?

Reviewing code changes for race conditions requires an adversarial edge-case pass that scales with diff size. This method probes modifications for concurrency issues and failure modes after initially verifying structural correctness and consistency.

Can I run a quick code review in my CI pipeline for local file changes?

Yes, you can run quick, standard, or thorough code review modes in CI pipelines for local file changes. The system auto-suggests the appropriate inspection intensity based on the diff size and whether sensitive paths are touched.

Does automated code review detect scope drift in feature work?

Automated code review detects scope drift by reading plan artifacts and comparing them against the current code changes. This ensures the modifications align with the intended feature work or bug fixes before producing a review artifact.

How do I audit code changes in security-sensitive paths like auth or payments?

Auditing code changes in security-sensitive paths like auth or payments uses a thorough inspection mode to catch design issues and regressions. It applies a two-pass review to identify structural bugs and security surface changes before shipping.

What are the limitations of automated code review for detecting complex bugs?

Automated code review focuses on structural correctness and adversarial edge cases rather than runtime behavior. It relies on git diffs or local file changes and operates within quick, standard, or thorough modes, which may not catch every logic flaw.