caveman
🪨 why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman.
- 114.5k stars
- 6.6k forks
- 96 open issues
- Language: Go
- License: Apache-2.0
- updated
Install
git clone https://github.com/JuliusBrussee/caveman- Clone the repository.
- Follow the README for installation and usage instructions.
Editor's note
Caveman is a Claude Code skill that instructs Claude and over 30 other AI coding agents to respond in compressed, telegraphic prose, stripping filler language while preserving technical accuracy. It installs via a single shell or PowerShell command and activates through a `/caveman` trigger, with four verbosity levels: `lite` (drops filler), `full` (default caveman speech), `ultra` (telegraphic), and `wenyan` (classical Chinese). Benchmarks against the Claude API show an average 65% reduction in output tokens across ten prompts, with individual tasks like explaining a React re-render bug hitting 87% reduction. Beyond compression, the skill includes `/caveman-commit` for Conventional Commit messages under 50 characters, `/caveman-review` for inline PR comments, `/caveman-compress` to rewrite context files like CLAUDE.md into caveman-speak (cutting roughly 46% of input tokens per session), and `caveman-shrink`, an MCP middleware that wraps any MCP server and compresses tool descriptions. Developers and teams paying per-token API costs are the primary beneficiaries, particularly those running long Claude Code sessions where output verbosity compounds expenses.
Tools overview
why many token when few do trick
Caveman make your AI agent say less and read less. Code stay exact. Brain still big.
33.2% fewerinput tokens, whole sessions54 Claude Code runs, 18 of 18 answers right 54% cheaperwith Auto, same score85 coding tasks vs Opus 5.5 at xhigh, interim 397 → 76tokens for one test runBlocks vs | tail -20 1.4 to 2.4× cheaperwith caveman-style outputAdobe Research, eight models
Cited by Adobe Research · A/B tested by JetBrains · Remade for Elasticsearch on Elasticsearch Labs #1 on Hacker News · #1 on GitHub Trending · "No way this actually works." ThePrimeagen
Install · Six rocks · Proof · Say less · Read less · Auto · Blocks · Commands · Caveman Cloud
Normal agent · 63 tokens Caveman agent · 20 tokens
The reason your React component is re-rendering is likely because you're creating a new object reference on each render cycle. When you pass an inline object as a prop, React's shallow comparison sees it as a different object every time, which triggers a re-render. I'd recommend using useMemo to memoize the object.
New object ref each render, so React re-renders. Wrap the prop in
useMemo.
Same fix. 63 token become 20. Brain still big.
Pick your club:
| Skill | Same answer | Tokens |
|---|---|---|
/caveman | New object ref each render, so React re-renders. Wrap the prop in useMemo. | 20 |
/ultracave | Inline object prop, new ref, re-render. useMemo. | 14 |
/megacave | 新參照致重繪。useMemo。 | 13 |
Token counts: tiktoken o200k.
Install
npx @caveman-ai/cli@latestOne command. Mac, Linux, Windows. Caveman find your agents and your logins, show what it change. Press Enter. Done. Needs Node.js 22.13+.
After that, plain claude stays caveman'd. So do Codex, Gemini CLI, OpenCode, and 30+ more. It runs on your machine, with your keys and your Claude Pro/Max login. One rock. That it.
Had caveman before 4.0? Say @latest, or the old copy on your machine wins. Already set up? caveman setup opens the screen again. Only want shorter answers? Get just the skill.
Six rocks, one install
New in 4.0: caveman is six parts. All on after install. Each has its own switch.
| Rock | What it does | Measured |
|---|---|---|
| Say less | Agent answers short. Code stays exact | 63 → 20 tokens, same fix |
| Read less | Logs, JSON, CSV, test output shrink before the model reads them | 33.2% fewer input tokens per session |
| Auto · new | Picks the model and the effort for each ask | 54% cheaper than Opus at xhigh, same score |
| Blocks · new | Code mode: agent keeps the scripts it writes and runs them again | 397 → 76 tokens for one test run |
| Waste fixes | Finds where your tokens go and fixes it. New: keeps the prompt cache warm while you pause | Not measured yet |
| Browse | Web pages come back compressed | 129.8× smaller page |
caveman status shows what is on. caveman off browse turns one off. Only Auto needs an account, and it is free. 4.0.0 release notes
Proof
Outside labs first. Ours come with each rock below. Nothing rounded up. Red rows stay red.
| Who | Setup | Result |
|---|---|---|
| Adobe Research, CAVEWOMAN paper (cites this repo) | Caveman-style output, eight models, five datasets | Cost cut 1.4 to 2.4× per model, up to 3× |
| Elastic, Elasticsearch Labs | Caveman mode remade for Elasticsearch, eight live MCP scenarios | 63.6% fewer response tokens. "Zero information loss." |
| JetBrains | 86 real coding tasks, paired A/B | No measurable quality loss (p = 0.82). 8.5% fewer output tokens |
Two findings shaped caveman. Adobe: cavemanning your prompt makes answers worse, so caveman never touches your prompt. JetBrains: agent sessions are mostly code and tool calls, which the voice leaves alone. So caveman grew rocks that shrink what the agent reads and what it runs on.
Say less: the caveman voice
A voice, not broken grammar. Every reply follows the same structure.
| Rule | What it means |
|---|---|
| Answer first | [thing] [action] [reason]. [next step]. No greeting, no "let me", no recap |
| One idea per sentence | Built on ASD-STE100, the controlled English of aircraft maintenance manuals: 20 words max, active voice, one term per thing |
| Meaning never dropped | Articles can go. not, never, no, only never go. Numbers and units stay exact |
| Payload verbatim | Code, commands, paths, and error messages untouched, character for character |
| No theater | No "me think". No chatter between tool calls. If caveman phrasing isn't shorter, plain wins |
| Knows when to stop | Security warnings, irreversible actions, and confused users get full sentences. Then grunt resumes |
| Your prompts stay yours | Never rewritten. Research say that backfire |
Ten dev questions on claude-opus-5-5:
| Instruction | Output tokens |
|---|---|
| None | 6,983 |
Answer concisely. | 4,334 |
/caveman | 4,119 |
/ultracave | 2,693 |
"Be concise" is the real baseline; new models already know it. On top of it, /caveman cuts 3% more at the median and /ultracave 35% more. Measured on the 3.1.0 rules, not yet re-run. The rules cost about 1,160 tokens of input, and if you pay per request (GitHub Copilot), shorter answers save nothing. Harness · Every case where caveman loses
Only want this rock? The skill alone, no proxy:
npx skills add JuliusBrussee/caveman -gType /caveman ($caveman in Codex) if it doesn't start on its own. Say stop caveman to go back.
Other ways in: Claude app and phone, Claude Code plugin, Gemini, every agent at once, Windows, uninstall
Claude on the web, the phone app, or Cowork? No terminal. Add caveman to your Claude account once, in Customize > Plugins: steps. Not tested end to end yet.
# Claude Code plugin, auto-starts every interactive session, subagents too
claude plugin marketplace add JuliusBrussee/caveman && claude plugin install caveman@caveman
# Gemini CLI
gemini extensions install https://github.com/JuliusBrussee/caveman
# Every agent on your machine at once, plus the Claude Code statusline badge (Node.js 22.13+)
curl -fsSL https://raw.githubusercontent.com/JuliusBrussee/caveman/v4.0.1/install.sh | bashAlready inside Claude Code? Same rock, slash form. Type these where you talk to Claude:
/plugin marketplace add JuliusBrussee/caveman
/plugin install caveman@cavemanWindows, PowerShell 5.1+:
irm https://raw.githubusercontent.com/JuliusBrussee/caveman/v4.0.1/install.ps1 | iexOn npm 12 or newer, new npm block git install. One-liners above handle it themselves from v3.2.0 on. Old one-liner pinned to v3.1.0 or older still fail there: use one above, or this:
npx --allow-git=root -y github:JuliusBrussee/caveman#v4.0.1
Changed your mind: npx -y github:JuliusBrussee/caveman -- --uninstall (on npm 12+, add --allow-git=root too)
Only grabbed the skill? npx skills remove JuliusBrussee/caveman -g take it back out.
What people ask about caveman
What is JuliusBrussee/caveman?
JuliusBrussee/caveman is tools for the Claude AI ecosystem. 🪨 why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman. It has 114.5k GitHub stars and its last recorded update is dated 2026-10-23.
How do I install caveman?
You can install caveman by cloning the repository (https://github.com/JuliusBrussee/caveman) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is JuliusBrussee/caveman safe to use?
Our security agent has analyzed JuliusBrussee/caveman and assigned a Trust Score of 97/100 (tier: Verified). See the full breakdown of passed checks and flags on this page.
Who maintains JuliusBrussee/caveman?
JuliusBrussee/caveman is maintained by JuliusBrussee. The last recorded GitHub activity is dated 2026-10-23, with 96 open issues.
Are there alternatives to caveman?
Yes. On ClaudeWave you can browse similar tools at /categories/tools, sorted by popularity or recent activity.
Maintain this repo? Show its Trust Score in your README
The badge shows this repository's current Trust Score and verdict, and links back to this page. It follows the catalogue data: nobody can buy or edit the score it displays.
[](https://claudewave.com/en/repo/juliusbrussee-caveman)<a href="https://claudewave.com/en/repo/juliusbrussee-caveman"><img src="https://claudewave.com/api/badge/juliusbrussee-caveman" alt="ClaudeWave Trust Score for JuliusBrussee/caveman" width="320" height="64"></a>