Bmad Review Adversarial General
- 324 installs
- 51.5k repo stars
- Updated August 5, 2026
- bmad-code-org/bmad-method
bmad-review-adversarial-general is a BMAD-METHOD core skill that performs cynical adversarial review on diffs, specs, stories, or documents and produces a findings report for developers who need skeptical pre-merge criti
About
bmad-review-adversarial-general is a bmad-code-org/BMAD-METHOD core skill that performs cynical adversarial review on any submitted artifact. The agent adopts a skeptical reviewer persona, loads content from a diff, branch, uncommitted changes, spec, story, or document, and executes a three-step flow: receive content, adversarial analysis, and present findings. The skill requires finding at least ten issues to fix or improve, looking for what is missing—not only what is wrong—and outputs a Markdown list of descriptions without severity rankings. bmad-code-review invokes this skill as the Blind Hunter layer during parallel code review. Developers reach for bmad-review-adversarial-general when they request a critical or cynical review of plans, specs, or implementations and need explicit weak-assumption surfacing before merge or release.
- Challenges assumptions with deliberate counterarguments
- Surfaces political, technical, and UX blind spots
- Works on PRDs, designs, or implementation plans
- Forces documented responses to serious objections
- Complements BMAD method structured review rituals
Bmad Review Adversarial General by the numbers
- 324 all-time installs (skills.sh)
- Ranked #275 of 1,352 Code Review & Quality skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/bmad-code-org/bmad-method --skill bmad-review-adversarial-generalAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 324 |
|---|---|
| repo stars | ★ 51.5k |
| Last updated | August 5, 2026 |
| Repository | bmad-code-org/bmad-method ↗ |
How do you cynically stress-test specs and diffs?
Stress-test plans, specs, or implementations by arguing opposing viewpoints, surfacing weak assumptions, and forcing explicit responses before merge or release.
Who is it for?
Developers and tech leads who want a cynical second opinion on diffs, specs, stories, or docs before approving merge or release decisions.
Skip if: Reviewers who need scored severity rankings, automated fix patches, or lightweight encouragement-focused feedback instead of skeptical findings-only output.
When should I use this skill?
A user requests a critical review, cynical review, or adversarial review of a diff, spec, story, document, or implementation artifact.
What you get
Markdown adversarial findings list with at least ten issues covering gaps, weak assumptions, and contradictions in the reviewed artifact.
- Markdown adversarial findings list
By the numbers
- Requires at least 10 adversarial findings per review
- 3 execution steps: receive content, adversarial analysis, present findings
Files
Adversarial Review (General)
Goal: Cynically review content and produce findings.
Your Role: You are a cynical, jaded reviewer with zero patience for sloppy work. The content was submitted by a clueless weasel and you expect to find problems. Be skeptical of everything. Look for what's missing, not just what's wrong. Use a precise, professional tone — no profanity or personal attacks.
Inputs:
- content — Content to review: diff, spec, story, doc, or any artifact
- also_consider (optional) — Areas to keep in mind during review alongside normal adversarial analysis
EXECUTION
Step 1: Receive Content
- Load the content to review from provided input or context
- If content to review is empty, ask for clarification and abort
- Identify content type (diff, branch, uncommitted changes, document, etc.)
Step 2: Adversarial Analysis
Review with extreme skepticism — assume problems exist. Find at least ten issues to fix or improve in the provided content.
Step 3: Present Findings
Output findings as a Markdown list: descriptions only, no severity, priority, or ranking.
HALT CONDITIONS
- HALT if zero findings — this is suspicious, re-analyze or ask for guidance
- HALT if content is empty or unreadable
Related skills
Improve Codebase ArchitectureSafely deepen clusters of shallow modules into cohesive, testable units while respecting their external dependencies.566k196k
Caveman ReviewGet ultra-compressed, one-line code review comments that cut noise while keeping every actionable fix.270k94.7kCodebase DesignShared vocabulary for designing deep modules: improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable.266k196kCavecrewDelegate coding tasks to specialized subagents that return compressed output, keeping the main context window usable for much longer sessions.220k94.7kCode ReviewReviews a branch or PR diff on two axes at once: conformance to coding standards plus a code-smell baseline, and whether it actually implements the original spec.201k196k
Requesting Code ReviewDispatch a consistent, high-signal code reviewer subagent that catches plan deviations and quality issues before merging or continuing development.183k264kHow it compares
Use bmad-review-adversarial-general for cynical findings-only critique; choose bmad-code-review when you need full parallel review orchestration with triage and Acceptance Auditor integration.
FAQ
How many findings must bmad-review-adversarial-general produce?
bmad-review-adversarial-general requires finding at least ten issues to fix or improve in the submitted content. If zero findings are returned, the skill halts because an empty result is treated as suspicious and triggers re-analysis.
What inputs does bmad-review-adversarial-general accept?
bmad-review-adversarial-general reviews diffs, branches, uncommitted changes, specs, stories, and documents passed as content, with an optional also_consider field for extra review focus areas alongside standard adversarial analysis.