bmad-code-review

Identify and categorize code-quality issues in changes through parallel adversarial review layers.

3|2|Updated Mar 29, 2024
One-click install
npx skills add https://github.com/josemariafs/MVTools --skill bmad-code-review-josemariafs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-code-review
Source: https://github.com/josemariafs/MVTools/tree/main/.cursor/skills/bmad-code-review
Command: npx skills add https://github.com/josemariafs/MVTools --skill bmad-code-review-josemariafs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Review code changes adversarially using parallel review layers (Blind Hunter, Edge Case Hunter, Acceptance Auditor) with structured triage into actionable categories. Use when the user says "run code review" or "review this code".

Core Features & Use Cases

  • Parallel reviewers: Blind Hunter, Edge Case Hunter, and Acceptance Auditor run in tandem to surface diverse issues.
  • Structured triage: findings are normalized, deduplicated, and classified into patch, decision_needed, defer, or dismiss.
  • Sprint-aligned output: integrates with story and sprint tracking and can write results to a story file.

Quick Start

Command the skill to perform a parallel, adversarial review on the current diff.

Frequently Asked Questions about bmad-code-review

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

FAQPage Schema
How do I automate adversarial code review for pull requests and diffs?

Adversarial code review automates pull request analysis by running parallel reviewers—Blind Hunter, Edge Case Hunter, and Acceptance Auditor—to surface and categorize diverse code-quality issues from your diffs.

What is the best way to triage code review findings into actionable sprint tasks?

Triage code review findings by normalizing and deduplicating them, then classifying each issue into patch, decision_needed, defer, or dismiss categories to integrate directly into your sprint tracking.

Can I run parallel review modes on specific commits and branches?

Yes, you can run parallel review modes on specific commits and branches by constructing a diff from those changes, which the reviewers then analyze for code-quality issues and edge cases.

How does an adversarial code review workflow handle edge case detection?

Adversarial code review handles edge case detection through the Edge Case Hunter reviewer, which runs in tandem with other layers to specifically target and surface hidden boundary issues within your code changes.

Does this code review automation require dependencies to function?

No, this code review automation requires no external dependencies to function, operating independently to perform diff analysis, structured triage, and sprint integration for your code changes.

When should I use an adversarial diff analysis instead of a standard code review?

Use adversarial diff analysis when you need to surface diverse, hidden code-quality issues that standard reviews miss, utilizing parallel blind hunting, edge case detection, and acceptance auditing for comprehensive coverage.