bmad-code-review

Run parallel adversarial code review layers on git diffs and triage findings.

Updated May 5, 2026
One-click install
npx skills add https://github.com/devalexanderdaza/bmad-any --skill bmad-code-review-devalexanderdaza
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-code-review
Source: https://github.com/devalexanderdaza/bmad-any/tree/main/.agents/skills/bmad-code-review
Command: npx skills add https://github.com/devalexanderdaza/bmad-any --skill bmad-code-review-devalexanderdaza

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Standard code reviews often miss subtle bugs, edge cases, and spec misalignments, leading to merged defects and rework. This Skill eliminates that gap by running multiple adversarial review layers in parallel and triaging findings into clear, actionable categories so you only focus on what matters.

Core Features & Use Cases

  • Parallel Adversarial Review Layers: Runs three specialized reviewers simultaneously: Blind Hunter (catches obvious issues with no project context), Edge Case Hunter (finds edge cases with full project access), and Acceptance Auditor (validates changes against provided specs and context docs).
  • Structured Triage: Normalizes, deduplicates, and classifies all findings into decision-needed, patch, defer, or dismiss buckets, removing noise and prioritizing actionable items.
  • Sprint Integration: Automatically syncs review outcomes to sprint status files and updates story status, fitting seamlessly into BMAD agile workflows.
  • Use Case: Ideal for teams running BMAD workflows who need consistent, thorough pre-merge code reviews that align with sprint requirements and catch issues before they reach production.

Quick Start

Use the bmad-code-review skill to review the uncommitted changes in your current feature branch against the main branch.

Frequently Asked Questions about bmad-code-review

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

FAQPage Schema
How do I run an adversarial code review on my git diff?

Adversarial code review runs parallel review layers on your git diff to catch bugs and spec violations that single-pass reviews miss. This skill automates that process using Blind Hunter, Edge Case Hunter, and Acceptance Auditor reviewers.

What is adversarial code review and how does it improve pre-merge validation?

Adversarial code review is a technique applying multiple specialized reviewers simultaneously to find edge cases and spec misalignments. It improves pre-merge validation by triaging findings into decision-needed, patch, defer, or dismiss buckets to remove noise.

Can I use this code review skill for large diffs over 3000 lines?

Yes, you can use this code review skill for large diffs over 3000 lines because it supports chunking. It processes large git diff outputs systematically to ensure comprehensive edge case analysis and spec compliance.

Does this code review tool require spec files for acceptance auditing?

Spec files are optional but recommended for acceptance auditing. Providing spec files and context docs allows the Acceptance Auditor layer to validate your code changes against sprint requirements and project specifications.

How do I sync code review outcomes to my BMAD sprint tracking workflow?

To sync code review outcomes to BMAD sprint tracking, the skill automatically updates sprint status files and story status after completing the review. This integration ensures pre-merge validation results align directly with your agile workflow.

What is the best way to triage code review findings for feature branch validation?

The best way to triage code review findings is to normalize, deduplicate, and classify them into actionable categories. This skill automates triage by sorting issues into decision-needed, patch, defer, or dismiss buckets for feature branch validation.