adversarial-review

Fresh adversarial code review with binary PASS/FAIL verdicts, evidence citations, and anchoring bias prevention via fresh reviewer spawning.

509 stars

Best use case

adversarial-review is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Fresh adversarial code review with binary PASS/FAIL verdicts, evidence citations, and anchoring bias prevention via fresh reviewer spawning.

Teams using adversarial-review should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/adversarial-review/SKILL.md --create-dirs "https://raw.githubusercontent.com/a5c-ai/babysitter/main/library/methodologies/metaswarm/skills/adversarial-review/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/adversarial-review/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How adversarial-review Compares

Feature / Agentadversarial-reviewStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Fresh adversarial code review with binary PASS/FAIL verdicts, evidence citations, and anchoring bias prevention via fresh reviewer spawning.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

Related Guides

SKILL.md Source

# Adversarial Review

## Overview

Independent adversarial code review checking spec compliance. Uses binary PASS/FAIL verdicts (not subjective feedback) with required file:line evidence citations.

## When to Use

- After quality gates pass in the execution loop
- For final comprehensive cross-unit review
- When verifying spec compliance of any implementation

## Key Differences from Collaborative Review

| Aspect | Collaborative | Adversarial |
|--------|--------------|-------------|
| Goal | Help improve code | Verify spec compliance |
| Verdict | Suggestions | Binary PASS/FAIL |
| Evidence | Optional | Required (file:line) |
| Reviewer | Can be reused | Must be fresh |
| Context | Shared | Independent |

## Fresh Reviewer Rule

On re-review after FAIL, a NEW reviewer instance spawns with no memory of the previous review. This prevents anchoring bias where a reviewer fixates on previously identified issues.

## Anti-Patterns

- Reusing reviewers after FAIL
- Passing previous findings to new reviewers
- Providing subjective or advisory feedback
- Accepting partial compliance as PASS

## Tool Use

Invoke as part of: `methodologies/metaswarm/metaswarm-execution-loop` (Phase 3)

Related Skills

systematic-review

509
from a5c-ai/babysitter

Conduct comprehensive literature searches, quality assessments, evidence synthesis, and meta-analyses

quality-assurance-review

509
from a5c-ai/babysitter

Conduct systematic quality reviews of instructional materials using established rubrics (Quality Matters) and design standards

peer-review-simulator

509
from a5c-ai/babysitter

Skill for simulating peer review feedback on manuscripts

dfm-review

509
from a5c-ai/babysitter

Skill for design for manufacturing review and optimization

design-review

509
from a5c-ai/babysitter

Skill for formal design review preparation and execution (PDR/CDR)

design-review-facilitator

509
from a5c-ai/babysitter

Design review planning and execution skill for structured design phase gate reviews per 21 CFR 820.30

clinical-literature-reviewer

509
from a5c-ai/babysitter

Systematic literature review skill for clinical evaluation supporting regulatory submissions

performance-review

509
from a5c-ai/babysitter

Generate performance review documentation and facilitate evaluation processes

requesting-code-review

509
from a5c-ai/babysitter

Use when completing tasks, implementing major features, or before merging to verify work meets requirements.

receiving-code-review

509
from a5c-ai/babysitter

Use when receiving code review feedback, before implementing suggestions. Requires technical rigor and verification, not blind implementation.

security-review

509
from a5c-ai/babysitter

Security vulnerability assessment identifying OWASP risks, injection vectors, authentication issues, and data exposure with severity classification.

plan-review-gate

509
from a5c-ai/babysitter

Adversarial plan review by 3 independent reviewers (Feasibility, Completeness, Scope & Alignment) before presenting to user.