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.
npx skills add oalders/kitchen-sink --skill adversarial-reviewgit clone --depth 1 https://github.com/oalders/kitchen-sinkWrote 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.
[](https://agentmods.dev/skills/oalders/kitchen-sink/adversarial-review)<a href="https://agentmods.dev/skills/oalders/kitchen-sink/adversarial-review"><img src="https://agentmods.dev/badge/skills/oalders/kitchen-sink/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.
<a href="https://agentmods.dev/skills/oalders/kitchen-sink/adversarial-review"><img src="https://agentmods.dev/badge/skills/oalders/kitchen-sink/adversarial-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00052 | $0.02026 |
| Opus 5 | $0.00026 | $0.01013 |
| Sonnet 5 | $0.00010 | $0.00405 |
| Haiku 4.5 | $0.00005 | $0.00203 |
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.
How it starts
The opening of the file, as written. The whole thing — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Adversarial Review
Two subagents review the same work in parallel, competing under an incentive that rewards in-scope defects with working repros and penalizes nit-padding and feature-proposal drift. The skill enforces scope discipline up front and triages findings against scope before presenting them.
The failure mode this skill prevents: without scope, reviewers find "everything" — theoretical contract gaps, "what if a future caller does X" cases, POD nits — and fixes accumulate surface that the next round attacks. Findings count stays high while real signal converges to zero.
When to Use
- The user asks for an "adversarial review", "review this adversarially", or wants two reviewers competing.
- You have a concrete patch, PR, plan, or change to review (not an open-ended question).
- You can already name the invariants the patch claims and its threat model (the precondition Step 1 enforces).
If you can't supply the preamble, this is the wrong skill. Use superpowers:requesting-code-review or /request-review instead — those don't require scope to work.
Required Inputs
The caller MUST provide all of the following. If any is missing, abort using the Step 1 message (which Step 2 also reuses for the round-number case) — do not improvise defaults, do not dispatch with placeholders.
- Scope preamble (Step 1): invariants, threat model, out-of-scope list.
- Round number (Step 2): positive integer counting this invocation.
- Work under review: a patch ref, diff, or file paths.
Step 1: Scope preamble (hard gate)
Before dispatching, the caller MUST supply:
- Invariants (1-3, crisp): what does the patch claim to make true?
- Threat model: who is the attacker, what's their leverage, what's the impact target?
- Out-of-scope list: which kinds of findings are not in scope? (e.g. "feature proposals", "theoretical contract gaps", "documentation wording")
If the caller can't supply these, abort with this exact message:
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.
- 11d ago First seen · 176 lines · 52 tokens per session scan A 3a93b1eb144a
adversarial-review is a skill published in the GitHub repository oalders/kitchen-sink (4 stars, last pushed 11d ago), licensed MIT. It adds 52 tokens to every session and 2,026 once invoked, about $0.0003 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.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…