bmad-review

Reviews diffs, code, and documents through configurable adversarial, edge-case, verification-gap, and editorial lenses.

Updated Jul 2, 2026
One-click install
npx skills add https://github.com/tuanpa-nhg-eng/nhg-ipms --skill bmad-review-tuanpa-nhg-eng
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-review
Source: https://github.com/tuanpa-nhg-eng/nhg-ipms/tree/main/.claude/skills/bmad-review
Command: npx skills add https://github.com/tuanpa-nhg-eng/nhg-ipms --skill bmad-review-tuanpa-nhg-eng

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Code and document reviews are inconsistent: reviewers miss edge cases, skip verification checks, or give unstructured feedback. This Skill runs a structured multi-lens review over any diff, file, spec, or document and returns findings in one canonical format. ## Core Features & Use Cases - Multi-lens review: Shipped lenses cover adversarial analysis, edge-case hunting, verification-gap detection, document structure, and prose copy-editing, each with its own method and findings shape. - Configurable workflow: A customize.toml file controls which lenses run, output format (JSON, markdown, or both), style guides, reader calibration, and standing review directives, with team and personal override files. - Structured findings output: Every finding carries location, trigger condition, guard snippet, and potential consequence, with no severity ranking, and can be written to a report file or presented in chat. - Use Case: Ask for a critical review of a pull request diff; the edge-case lens enumerates unhandled branches, the verification-gap lens checks whether tests would catch a regression, and results arrive as a grouped markdown report plus raw JSON. ## Quick Start Review the changes in my current git diff with all applicable lenses and show me the findings.

Frequently Asked Questions about bmad-review

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

FAQPage Schema
How do I review a git diff for edge cases automatically?

Run the review with the edge-case-hunter lens on your diff. It exhaustively walks every branch and boundary condition in the changed hunks, reports only unhandled paths, and returns findings as JSON with location, trigger condition, guard snippet, and consequence.

How do I check whether my code change has test coverage gaps?

Use the verification-gap lens, which traces changed behavior to its consumers and reads the relevant tests. It reports regression gaps, missing-adoption gaps, and broken-verification gaps only when backed by evidence from tests it actually inspected.

Can I customize which review lenses run and the output format?

Yes. The customize.toml file defines the lens set, output format (json, markdown, or both), style guide, and report path. Team and personal override TOML files merge per BMad rules, and you can also request specific lenses by name at invocation time.

Does the editorial review change my document's content?

No. The structure and prose lenses treat content as sacrosanct and only critique organization and expression. They return a findings table of suggested cuts, merges, and copy-edits that the author accepts or rejects row by row.

What happens when a review finds no issues?

An empty findings array is a valid result for most lenses, and the markdown report states the review is clean. The adversarial lens is the exception: it requires at least ten findings and treats an empty list as a signal to re-check.