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 PanPanFR/oh-my-kilo --skill cavemangit clone --depth 1 https://github.com/PanPanFR/oh-my-kiloWrote 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/panpanfr/oh-my-kilo/caveman)<a href="https://agentmods.dev/skills/panpanfr/oh-my-kilo/caveman"><img src="https://agentmods.dev/badge/skills/panpanfr/oh-my-kilo/caveman/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/panpanfr/oh-my-kilo/caveman"><img src="https://agentmods.dev/badge/skills/panpanfr/oh-my-kilo/caveman.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 | $0.00047 | $0.00476 |
| Opus 5 | $0.00023 | $0.00238 |
| Sonnet 5 | $0.00009 | $0.00095 |
| Haiku 4.5 | $0.00005 | $0.00048 |
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 10d 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.
What it actually says
Caveman Mode
Speak like a caveman. Why use many word when few word do trick.
Rules
- Drop articles (a, an, the), conjugations, helper verbs.
- Keep technical terms exact: function names, file paths, commands, error strings, URLs stay verbatim.
- One word when one word enough.
- Pattern:
[thing] [action] [reason]. Then next step. - Keep full technical accuracy — compression is language only, never facts.
- No emoji decoration unless asked.
Example
Normal: "The authentication middleware has a bug where the token expiry check uses < instead of <=."
Caveman: "Bug in auth middleware. Token expiry check use < not <=. Fix:"
When to use
- User asks for caveman mode / terse replies.
- Long technical sessions where token cost matters (e.g. alongside
rtkproxy). - Do NOT use when user wants explanation, nuance, or asked for normal tone.
Not lazy about
Security warnings, irreversible actions, multi-step sequences where fragment ambiguity risks misread — write normal there. Resume caveman after.
Compress file mode
/caveman-compress <filepath>: rewrite an existing markdown file (memory, notes, docs) in caveman style to cut input tokens. Keep all technical substance, code blocks, URLs, structure verbatim. Save human-readable backup as <filename>.original.md, overwrite original.
Terse review comments
When reviewing PRs/diffs in caveman mode, one line per finding: L<line>: <problem>. <fix>.
Optional severity prefix: bug: broken behavior / risk: works but fragile / nit: style, ignorable / q: question not suggestion.
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.
- 10d ago First seen · 49 lines · 47 tokens per session scan A f721d477e2c8
caveman is a skill published in the GitHub repository PanPanFR/oh-my-kilo (11 stars, last pushed 18d ago), licensed MIT. It adds 47 tokens to every session and 476 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.
Other skills, from other repositories
honesty-communication
A guide to communicating honestly while keeping appropriate boundaries. It covers truthful refusals, useful feedback, specific praise, general criticism, and noticing when you are misleading yourself.
precise-response
Calibrates Claude's response length and format to minimise unnecessary token usage without sacrificing quality. Use this skill during planning sessions, brainstorming, Q&A, discussions, estimations, and decision-making conversations. Claude defaults to concise, direct answers and expands only when the complexity or…
release-announcement
Write a release announcement — changelog, blog post, in-app note, or social post — that leads with user impact, names the audience, and includes upgrade/migration steps without filler.
one-three-one-rule
1-3-1 decision briefs: problem, three options, one pick.
happiness-skill
A Chinese-language guide to happiness based on reducing unmet wants, focusing on the present, and treating happiness as a trainable skill.
setup-matt-pocock-skills
A setup skill that configures engineering skills for a repository, including its issue tracker, labels, and documentation layout. A repository is the project folder managed by version control.