adversarial-review

adversarial-review is a skill for Claude Code from AlexShchuka/neuro-matrix. It costs 72 tokens per session (1,510 once invoked), scanned A, original, MIT.

An independent audit step for work produced by several AI agents before it is merged, deployed, or used in the next phase. It treats the reviewer as someone looking for ways the system can fail.

In plain words
What is it for?
Use it after a multi-agent build to produce prioritized findings and identify blockers requiring human review.
Why use it?
It helps expose security problems, brittle behavior, and incorrect assumptions that a friendly review may overlook.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the neuro-matrix plugin — 6 skills, 5 agents, 4 hooks shipped together

Good fit Use it after a multi-agent build to produce prioritized findings and identify blockers requiring human review.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/alexshchuka/neuro-matrix/adversarial-review
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 AlexShchuka/neuro-matrix --skill adversarial-review
Clone the repo
git clone --depth 1 https://github.com/AlexShchuka/neuro-matrix

Made for: Claude Code.

Or install neuro-matrix, the plugin that ships this one along with the rest of its 6 skills, 5 agents, 4 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-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/alexshchuka/neuro-matrix/adversarial-review/github.svg)](https://agentmods.dev/skills/alexshchuka/neuro-matrix/adversarial-review)
Your own site
<a href="https://agentmods.dev/skills/alexshchuka/neuro-matrix/adversarial-review"><img src="https://agentmods.dev/badge/skills/alexshchuka/neuro-matrix/adversarial-review/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-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/alexshchuka/neuro-matrix/adversarial-review"><img src="https://agentmods.dev/badge/skills/alexshchuka/neuro-matrix/adversarial-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 72 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,510 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00072 $0.01510
Opus 5 $0.00036 $0.00755
Sonnet 5 $0.00014 $0.00302
Haiku 4.5 $0.00007 $0.00151

Measured 11d ago against content hash 462889d18e94, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

adversarial-review scanned grade A with 0 findings 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 11d ago.

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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/adversarial-review/SKILL.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Adversarial-review: an independent audit gate, not a helper

BLUF: the dominant failure mode of post-build review is sycophancy — the reviewer treats «looks plausible» as confirmation, confirms what the author implied was correct, and produces approval theater. This skill makes that structurally impossible by framing the reviewer as an adversary whose job is to break the work, not bless it.

Stage 0 — frame the reviewer correctly

The reviewer is an independent external auditor (mental model: PhD architect encountering this code for the first time with no social obligation to the author). The instructions to the reviewer must include:

  • Do NOT confirm. Do NOT hint. Do NOT ask leading questions. Do NOT soften.
  • Your job is to find every way this system fails, is insecure, is brittle, or is wrong.
  • «I could not find a problem in area X» is an admissible finding. «Everything looks good» is not.
  • Every factual claim requires ≥2 confirmations: a code anchor (file:line or exact snippet) PLUS an external reference (spec, RFC, reference repo, web) OR an explicit «cannot verify without execution» with a stated reason.

Do NOT ask the reviewer to «confirm this is working» or «verify the implementation is correct» — those framings produce sycophancy. Ask: «find every blocker that would prevent this from working correctly in production.»

Stage 1 — audit dimensions (fixed; do not abbreviate)

Run the reviewer across all dimensions in a single pass. Skipping a dimension is only valid if explicitly stated with a reason.

Dimension What to look for
Concurrency / lifecycle data races, leaked goroutines/threads, deadlocks, use-after-free, resources not GC-safe (closed by GC rather than owned lifecycle), subscriptions re-created per-event instead of once per session
Error handling swallowed errors, false success (operation failed but caller sees success), missing propagation, error type that loses context
Isolation / security credentials leaking across trust boundaries, over-privileged components, missing input validation at trust boundaries
Architecture & code smell duplicated logic, reinvented wheels (stdlib or idiomatic library solves this), wrong layer (business logic in transport, I/O in domain), dead code with a live twin (two sources of truth)
Benchmark vs reference repos compare idiom usage to 2–3 reference repos on the same problem; flag where this code diverges from the community's solved pattern
Tests-as-theatre tests that always pass regardless of behavior, mocks that make the test vacuous, test coverage of the wrong layer
Robustness bar every I/O and wait: is there a hard timeout/deadline? is there a forced unblock if the far side hangs? does the system degrade gracefully or panic?

Read the full file on GitHub · 78 lines

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. 11d ago First seen · 78 lines · 72 tokens per session scan A 462889d18e94

Subscribe to this mod's changes

adversarial-review is a skill published in the GitHub repository AlexShchuka/neuro-matrix (3 stars, last pushed 2mo ago), licensed MIT. It adds 72 tokens to every session and 1,510 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

md-audit

Read-only code quality audit — scan the current working directory for common issues (bugs, dead code, security hotspots, missing error handling) and return a prioritised findings report. No files are edited. Use when asked to "audit the code", "quick audit", "find issues", "code scan", or "what's wrong with this…

chaitanyagiri/munder-difflin · 85 tokens

external-pr-review

Apply the external contributor PR review procedure. Collect PR information, write mydocs/pr/pr{N}review.md, verify, write pr{N}report.md, and move documents to archives/ when processing is complete. Use only for external contributor PRs, not internal tasks.

postmelee/hyper-waterfall · 60 tokens

receiving-code-review

Processes code review feedback systematically by classifying findings, deciding fix or reject with evidence, applying approved fixes, and re-verifying before marking comments resolved.

tranhieutt/software_development_department · 35 tokens

code-review

Performs architectural and quality code review checking coding standards, SOLID principles, architectural compliance, and common software issues. Use when reviewing a file or directory before merge, or when the user mentions code review, PR review, or quality check.

tranhieutt/software_development_department · 51 tokens

tech-debt

Scans the codebase for technical debt indicators, categorizes and prioritizes findings, and maintains a debt register with repayment recommendations. Use when assessing codebase health, planning refactoring, or when the user mentions technical debt or code quality.

tranhieutt/software_development_department · 52 tokens

code-review-checklist

Provides a comprehensive code review checklist for pull requests covering security, performance, maintainability, and testing. Use as a reference during code reviews or when the user asks for a review checklist.

tranhieutt/software_development_department · 42 tokens