adversarial-reviewer

adversarial-reviewer is a skill for Claude Code from mhylle/claude-skills-collection. It costs 288 tokens per session (5,213 once invoked), scanned A, original, MIT.

A code-review process that asks three separate reviewers to examine the same code: one tries to break it, one reads it as a new developer, and one looks for security problems.

In plain words
What is it for?
Use it to review code changes from several independent perspectives and combine the findings by severity and repeated discoveries.
Why use it?
It reduces the risk that the original author and reviewer share the same blind spots or assumptions.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions CLAUDE.md; mentions subagents.

Part of the devflow plugin — 38 skills, 13 agents, 5 hooks shipped together

Good fit Use it to review code changes from several independent perspectives and combine the findings by severity and repeated discoveries.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/mhylle/claude-skills-collection/adversarial-reviewer
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add mhylle/claude-skills-collection --skill adversarial-reviewer
Clone the repo
git clone --depth 1 https://github.com/mhylle/claude-skills-collection

Made for: Claude Code.

Or install devflow, the plugin that ships this one along with the rest of its 38 skills, 13 agents, 5 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for adversarial-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/skills/mhylle/claude-skills-collection/adversarial-reviewer/github.svg)](https://agentmods.dev/skills/mhylle/claude-skills-collection/adversarial-reviewer)
Your own site
<a href="https://agentmods.dev/skills/mhylle/claude-skills-collection/adversarial-reviewer"><img src="https://agentmods.dev/badge/skills/mhylle/claude-skills-collection/adversarial-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for adversarial-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/skills/mhylle/claude-skills-collection/adversarial-reviewer"><img src="https://agentmods.dev/badge/skills/mhylle/claude-skills-collection/adversarial-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 288 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,213 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
SkillSpector: 1 finding, up to low

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • low Privilege Escalation · line 277
    Skill requests more permissions than appear necessary for its stated functionality. Review if elevated access is justified.
    Fix: Request only the minimum permissions required. Document why each permission is needed. Remove broad permissions like '*' or 'all'.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00288 $0.05213
Opus 5 $0.00144 $0.02606
Sonnet 5 $0.00058 $0.01043
Haiku 4.5 $0.00029 $0.00521

Measured 12d ago against content hash c288c27371c5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

adversarial-reviewer scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

The scan reads SKILL.md. This mod also ships 3 executable files (evals/fixtures/test-1-saboteur/cache.ts, evals/fixtures/test-2-new-hire/process.py, evals/fixtures/test-3-security/auth.js), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

`child_process`, raw SQL, `pickle.loads`, `yaml.load` — these are useful
skills/adversarial-reviewer/SKILL.md · 392 lines

How it starts

The opening of the file, as written. The whole thing — 392 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Adversarial Code Reviewer

Adversarial code review that forces genuine perspective shifts by dispatching three hostile reviewer personas as independent subagents. Each subagent gets only the code and its persona brief — not the history of why the code was written, not Claude's reasoning about it, not prior "looks good" conclusions. Findings are then severity-classified and cross-promoted when caught by multiple personas.

Why Subagents Matter Here

The whole point of this skill is breaking out of the self-review trap: when Claude reviews code it just wrote or just read, the reviewing context shares the producing context's mental model and blind spots. If all three personas ran in the same context, the "perspective shift" would be cosmetic — same weights, same priming, same assumptions.

Running each persona as an isolated subagent forces a real reset:

  • Each subagent starts with only the diff, the files, and its persona brief.
  • It cannot know what problem the author was trying to solve, what was tried and rejected, or what Claude already concluded.
  • It cannot be swayed by another persona's findings — the three proceed in parallel.
  • Synthesis happens only after all three have independently reported.

This is the mechanism. Preserve it.

When to Use

  • Before merging any PR — especially self-authored PRs with no human reviewer
  • After a long coding session — fatigue produces blind spots
  • When Claude just said "looks good" — if approval came easy, run this for a second opinion
  • On security-sensitive code — auth, payments, data access, API endpoints
  • When something feels off — trust the instinct

Usage

/adversarial-review                     # Review staged/unstaged changes (default)
/adversarial-review --diff HEAD~3       # Review last 3 commits
/adversarial-review --diff main...HEAD  # Review a feature branch vs main
/adversarial-review --file src/auth.ts  # Review a specific file in full
/adversarial-review --codebase          # Review the whole repo (from CWD)
/adversarial-review --codebase src/api  # Review a specific subtree

Read the full file on GitHub · 392 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 392 lines · 288 tokens per session scan A c288c27371c5

Subscribe to this mod's changes

adversarial-reviewer is a skill published in the GitHub repository mhylle/claude-skills-collection (18 stars, last pushed 10d ago), licensed MIT. It adds 288 tokens to every session and 5,213 once invoked, about $0.0014 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.