kube-agents: Skill for Claude Code

.agents/skills/review-adversarial/SKILL.md

review-adversarial is a skill for Claude Code, Codex from gke-labs/kube-agents. It costs 45 tokens per session (4,348 once invoked), scanned A, original, Apache-2.0.

A search workflow for Semantic Scholar, a database of academic papers, authors, citations, and related research.

In plain words
What is it for?
It is for searching papers by topic or exact terms, finding authors and citations, locating passages, and getting research recommendations.
Why use it?
It provides a structured way to find papers and supporting details without making inefficient or repeated API requests.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions subagents.

This is gke-labs/kube-agents's own configuration. It tells Claude Code and Codex how to work on kube-agents itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything kube-agents configures →

Reuse

Borrowing it

Nothing to install: this file belongs to gke-labs/kube-agents. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/gke-labs/kube-agents/main/.agents/skills/review-adversarial/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/gke-labs/kube-agents

Made for: Claude Code, Codex.

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 review-adversarial

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

Your own site · 80×15
<a href="https://agentmods.dev/skills/gke-labs/kube-agents/review-adversarial"><img src="https://agentmods.dev/badge/skills/gke-labs/kube-agents/review-adversarial.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 45 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,348 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. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
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 $0.00045 $0.04348
Opus 5 $0.00023 $0.02174
Sonnet 5 $0.00009 $0.00870
Haiku 4.5 $0.00005 $0.00435

Measured 8d ago against content hash 1b841ea13812, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

review-adversarial 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 8d 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.

.agents/skills/review-adversarial/SKILL.md · 303 lines

How it starts

The opening of the file, as written. The whole thing — 303 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Task

Given a diff range, find the defects in it and report only the ones you can defend. This skill is the review method itself and holds no plumbing: it is run by an author against their own branch before opening a pull request (the AGENTS.md requirement, wrapped by review-preflight), and by a reviewer against a pull request already open (.claude/commands/pr-review-batch.md, which wraps it in the GitHub-side work).

review-docs-drift is the companion pass, not part of this one. Angle H stops at the rules the diff visibly breaks and leaves the rest of the documentation question to that skill.

Procedure

1. Run this in a context that did not write the change

If you are the agent that just produced the diff, do not run the pass in the conversation that produced it. review-preflight is how you get a context that did not: it owns the plumbing, down to what to hand the fresh context, what to withhold from it, and what to do when your harness will not spawn one for the asking. A model reviewing work it has just justified is the weakest configuration there is: it is poorly calibrated about its own output, rates it higher than an outsider would, and the bias is worst on exactly the lines it got wrong. It is also more likely to "fix" something correct than to catch something broken, which is why step 5 exists and why step 6 will not let you edit on a hunch.

That is why both wrappers spend a subagent on it — .claude/commands/pr-review-batch.md for a pull request already open, .claude/commands/pr-preflight.md for the author's own branch.

Without a wrapper, the minimum is: hand the fresh context the repository, the diff range, and this file, and withhold your plan, your reasoning, and the intent sentence step 3 asks it to derive for itself. The gap between what you meant and what the diff says is the finding you cannot get any other way.

Read the full file on GitHub · 303 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. 8d ago Changed · +3 lines 1b841ea13812
  2. 11d ago First seen · 300 lines · 45 tokens per session scan A 53c414f24eb3

Subscribe to this mod's changes

review-adversarial is a skill published in the GitHub repository gke-labs/kube-agents (53 stars, last pushed today), licensed Apache-2.0. It adds 45 tokens to every session and 4,348 once invoked, about $0.0002 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 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens