bmad-code-review

Coordinate parallel review layers to surface edge cases and acceptance gaps in code changes.

Updated Apr 26, 2026
One-click install
npx skills add https://github.com/Wizarck/iguanatrader --skill bmad-code-review-wizarck
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-code-review
Source: https://github.com/Wizarck/iguanatrader/tree/main/skills/bmad-code-review
Command: npx skills add https://github.com/Wizarck/iguanatrader --skill bmad-code-review-wizarck

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables adversarial evaluation of code changes by coordinating parallel review layers (Blind Hunter, Edge Case Hunter, Acceptance Auditor) and structured triage to produce actionable findings, especially when users request code reviews like "run code review" or "review this code".

Core Features & Use Cases

  • Parallel review layers: orchestrates multi-layer scrutiny to surface edge cases and acceptance gaps.
  • Structured triage: classifies findings into actionable categories for quick remediation.
  • Deterministic workflow: uses a step-file architecture with explicit steps, state tracking, and append-only artifacts to ensure reproducible reviews.
  • Use Case: a developer submits a diff; the skill triggers the three subagents and consolidates their findings into a cohesive QA report for the story workflow.

Quick Start

Trigger the code-review workflow for the latest changes.

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 a pull request?

Adversarial code review on a pull request is executed by coordinating parallel review layers, such as Blind Hunter and Edge Case Hunter, to surface defects. The process uses a deterministic step-file architecture to orchestrate multi-layer scrutiny and generate a structured triage report.

What is a parallel triage in a code review workflow?

Parallel triage in a code review workflow classifies defects uncovered by multiple subagents into actionable categories for quick remediation. It consolidates findings from independent review layers into a cohesive QA report to highlight edge cases and acceptance gaps.

Can I use this code review skill on branches and commits across different projects?

Yes, you can apply this code review skill to diffs, commits, or branches across multiple projects. The multi-layer review architecture evaluates code changes universally by orchestrating parallel reviewer prompts to ensure comprehensive defect detection.

How do you uncover edge cases and acceptance gaps in a diff?

Edge cases and acceptance gaps in a diff are uncovered through an adversarial evaluation mechanism that applies parallel review layers. An Acceptance Auditor specifically checks for requirement gaps while an Edge Case Hunter identifies boundary condition failures.

Do I need a step-file architecture to perform structured code reviews?

A step-file architecture is required to ensure deterministic and reproducible code reviews. It provides explicit steps, state tracking, and append-only artifacts to orchestrate the parallel reviewer prompts needed for comprehensive triage classification.