bmad-review

Runs one or more installed review lenses — adversarial critique, edge cases, verification gaps, structure, prose — and reports triaged findings. Use when, and only when, the user asks you to review a diff, a pull request, or an artifact — code or documents, one or many — and actually says "review"; an explicit skill:bmad-review directive from another skill counts as that ask. A request to act on feedback from an earlier review is a change, not a review. Never invoke this uninvited, including on edits you just made.

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

85%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-architected orchestration skill: clear sequenced workflow with validation, strong progressive disclosure via just-in-time lens references, and concrete commands and output shapes. The main weakness is prose density in the Execution steps, where long conditional sentences could be broken into shorter, scannable units.

Suggestions

Break the long conditional sentences in Execution steps 2, 5, and 6 into shorter sentences or nested bullets so the branching logic is scannable rather than parsed from a run-on paragraph.

Surface the resolved default lens set inline (or in a short table) so the orchestration is partially actionable even before {workflow.lenses} resolves, reducing reliance on placeholders for the core path.

Move the large inline subagent constraint quote in steps 5 and 6 into a shared reference snippet referenced once, since it is repeated verbatim and inflates the body.

Dimension	Reasoning	Score
Conciseness	Information-dense and free of concept over-explanation (no 'what a diff is' padding), but several Execution steps are long run-on paragraphs packed with conditional clauses that could be tightened into shorter sentences or sub-bullets.	4 / 5
Actionability	Provides a concrete executable command (uv run resolve_customization.py ...), exact JSON fallback shapes for empty input, and a precise output schema; minor gaps remain because most runtime behavior is delegated to {workflow.*} placeholders and just-in-time lens files.	4 / 5
Workflow Clarity	A clear 7-step sequence with explicit validation checkpoints and feedback loops — empty/undecodable input handling, a resolve-customization failure fallback, and staged-claims isolation — covering error recovery concretely.	5 / 5
Progressive Disclosure	SKILL.md is a lean overview that loads lens reference files just-in-time ('load only what runs'), references are one level deep and real (e.g. references/lens-edge-case-hunter.md exists), and the bundle is cleanly split across references/ and scripts/ with clear navigation.	5 / 5
	Total	18 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, third-person description that crisply states what the skill does and precisely when to invoke it, with explicit boundary guidance that sharply lowers conflict risk. Trigger terms are natural and well-covered; the only risk is density, not deficiency.

Dimension	Reasoning	Score
Specificity	Names multiple concrete capabilities — adversarial critique, edge cases, verification gaps, structure, prose — and the concrete action of reporting triaged findings, giving comprehensive coverage of what the skill does.	5 / 5
Completeness	Explicitly answers both what (runs review lenses, reports triaged findings) and when (Use when the user asks to review a diff, PR, or artifact and actually says 'review'), with concrete trigger phrases and an explicit non-trigger boundary.	5 / 5
Trigger Term Quality	Includes the natural terms users say — 'review', 'diff', 'pull request', 'artifact', 'code', 'documents' — with synonyms and one-or-many variations, matching comprehensive natural-term coverage.	5 / 5
Distinctiveness Conflict Risk	Clear niche (review lenses) with distinct triggers and strong boundary guidance — 'Use when, and only when', 'A request to act on feedback...is a change, not a review', 'Never invoke this uninvited' — minimizing conflict risk.	5 / 5
	Total	20 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: bmad-code-org/BMAD-METHOD
Path: src/core-skills/bmad-review/SKILL.md
Commit: 67d876f

Reviewed: about 21 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.