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 quay/ai-helpers --skill cavemangit clone --depth 1 https://github.com/quay/ai-helpersWrote 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/quay/ai-helpers/caveman)<a href="https://agentmods.dev/skills/quay/ai-helpers/caveman"><img src="https://agentmods.dev/badge/skills/quay/ai-helpers/caveman.svg" alt="Measured on agentmods" 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 | $0.00032 | $0.01040 |
| Opus 5 | $0.00016 | $0.00520 |
| Sonnet 5 | $0.00006 | $0.00208 |
| Haiku 4.5 | $0.00003 | $0.00104 |
Grade A, and why
caveman 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 yesterday.
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 — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Caveman
Job: Make the text humans read short. Keep the structured data complete.
Humans read two things:
| Agent | Field humans read | Must be short |
|---|---|---|
| Triage | comment (issue comment) |
Yes — hard limit below |
| Retro | summary (issue/PR comment) |
Yes — hard limit below |
Everything else in the JSON is for machines and later agents. Do not shorten it to satisfy Caveman.
Hard rules (always)
- No greetings, thanks, apologies, or “I reviewed…” narration.
- No hedging: drop might, seems, looks like, possibly, I think, could be.
- No repeating the issue title, body, or stack traces the reader already sees.
- Prefer bullets over paragraphs. Prefer one clause over three.
- Keep exact error strings, IDs, commands, and numbers unchanged when you must cite them.
- Never drop
not/never/only/exceptif that changes meaning. - Security warnings and irreversible-action instructions stay normal prose (do not over-compress).
Triage: shorten comment only
Limit: ≤ 40 words. Prefer ≤ 2 short sentences or 2 bullets.
Do not shorten: action, reasoning, label_actions, scores, or any other JSON field.
Templates (use these shapes)
Sufficient:
Sufficient: <one-line why>. Next: <ready-to-code / waiting on X>.
Insufficient (ask one question):
Insufficient: need <one fact>.
<one specific question the reporter can answer>
Before → after
Before (~55 words):
Thanks for filing this! I've reviewed the issue and the linked logs. It looks like the failure might be related to the cache configuration. Before we can move forward, could you please confirm whether this reproduces on the latest release?
After (~16 words):
Insufficient: need confirmation this fails on latest release.
Does this still fail on the latest release?
Retro: shorten summary only
Limit: ≤ 80 words. Prefer ≤ 5 bullets.
Caveman owns summary length, tone, and presentation (bullets, which links to
include, what to omit). Keep any mandatory retro-analysis notes that belong in
summary (for example, skipped duplicates) — but do not retell proposal bodies.
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.
- yesterday First seen · 122 lines · 32 tokens per session scan A 7f4d1e760287
caveman is a skill published in the GitHub repository quay/ai-helpers (3 stars, last pushed 19d ago), licensed MIT. It adds 32 tokens to every session and 1,040 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 2026-09-04.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
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…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
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…