bmad-code-review

Automate adversarial code review with parallel review layers and structured triage.

15|2|Updated Apr 15, 2026
One-click install
npx skills add https://github.com/lorenzogm/bmad-ui --skill bmad-code-review-lorenzogm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-code-review
Source: https://github.com/lorenzogm/bmad-ui/tree/main/.cursor/skills/bmad-code-review
Command: npx skills add https://github.com/lorenzogm/bmad-ui --skill bmad-code-review-lorenzogm

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a structured, adversarial code review process, improving code quality and developer efficiency by identifying and categorizing issues before deployment.

Core Features & Use Cases

  • Adversarial Review Layers: Utilizes parallel review layers including Blind Hunter, Edge Case Hunter, and Acceptance Auditor.
  • Structured Triage: Organizes findings into actionable categories like decision_needed, patch, defer, and dismiss.
  • Workflow Automation: Automates the code review process with step-file architecture and state tracking.
  • Use Case: Ideal for teams using BMAD (Build Manage Analyze Deploy) to streamline the code review process.

Quick Start

Run the 'bmad-code-review' skill and follow the prompts to initiate a code review.

Frequently Asked Questions about bmad-code-review

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

FAQPage Schema
What is adversarial code review and how does structured triage work?

Adversarial code review deploys parallel review layers like Blind Hunter and Edge Case Hunter to find issues. Structured triage then categorizes findings into decision_needed, patch, defer, or dismiss to organize code changes before deployment.

How do I automate code review for BMAD workflows?

You can automate code review for BMAD workflows by running the bmad-code-review skill and following the prompts. It applies step-file architecture and state tracking to execute parallel review layers without manual orchestration.

Can I use structured triage to categorize code changes before deployment?

Yes, structured triage organizes code review findings into actionable categories including decision_needed, patch, defer, and dismiss. This ensures issues are properly categorized and handled before deployment.

Does bmad-code-review require project configuration and source code access?

Yes, bmad-code-review requires project configuration and access to source code to perform its parallel review layers. You must have these prerequisites configured before initiating the automated review process.

What is the best way to identify edge cases and acceptance criteria issues in code?

Using parallel adversarial review layers like Edge Case Hunter and Acceptance Auditor is the best way to identify these issues. This approach catches edge cases and verifies acceptance criteria before changes reach deployment.