caveman

why use many token when few do trick

Before/After • Install • Levels • Skills • Benchmarks • Evals

🪨 Caveman Ecosystem  ·  caveman talk less _{(you are here)}  ·  cavemem remember more  ·  cavekit build better

---

A Claude Code skill/plugin and Codex plugin that makes agent talk like caveman — cutting ~75% of output tokens while keeping full technical accuracy. Now with 文言文 mode, terse commits / one-line reviews / lifetime stats, and a compression tool that cuts ~46% of input tokens every session.

Based on the viral observation that caveman-speak dramatically reduces LLM token usage without losing technical substance. So we made it a one-line install.

Before / After

🗣️ Normal Claude (69 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."	🪨 Caveman Claude (19 tokens) "New object ref each render. Inline object prop = new ref = re-render. Wrap in useMemo."
🗣️ Normal Claude "Sure! I'd be happy to help you with that. The issue you're experiencing is most likely caused by your authentication middleware not properly validating the token expiry. Let me take a look and suggest a fix."	🪨 Caveman Claude "Bug in auth middleware. Token expiry check use < not <=. Fix:"

Same fix. 75% less word. Brain still big.

Pick your level of grunt:

🪶 Lite "Your component re-renders because you create a new object reference each render. Inline object props fail shallow comparison every time. Wrap it in useMemo."

🪨 Full "New object ref each render. Inline object prop = new ref = re-render. Wrap in useMemo."

🔥 Ultra "Inline obj prop → new ref → re-render. useMemo."

📜 文言文 "物出新參照，致重繪。useMemo Wrap之。"

Same answer. You pick how many word.

┌─────────────────────────────────────┐
│  TOKENS SAVED          ████████ 75% │
│  TECHNICAL ACCURACY    ████████ 100%│
│  SPEED INCREASE        ████████ ~3x │
│  VIBES                 ████████ OOG │
└─────────────────────────────────────┘

• Faster response — less token to generate = speed go brrr
• Easier to read — no wall of text, just answer
• Same accuracy — all technical info kept, only fluff dropped (science say so)
• Save money — 65% mean output reduction across our benchmarks (range 22-87%)
• Fun — every code review become comedy

Install

One line. Detect every agent. Install for each.

# macOS / Linux / WSL / Git Bash
curl -fsSL https://raw.githubusercontent.com/JuliusBrussee/caveman/main/install.sh | bash

# Windows (PowerShell)
irm https://raw.githubusercontent.com/JuliusBrussee/caveman/main/install.ps1 | iex

Detects 30+ agents (Claude Code, Gemini CLI, Codex, Cursor, Windsurf, Cline, Copilot, Continue, Kilo, Roo, Augment, Aider Desk, Amp, Bob, Crush, Devin, Droid, ForgeCode, Goose, iFlow, JetBrains Junie, Kiro CLI, Mistral Vibe, OpenHands, opencode, Qwen Code, Qoder, Rovo Dev, Tabnine, Trae, Warp, Replit Agent, Antigravity, …). Runs each one's native install. Skips what you not have. Safe to re-run.

By default the installer wires Claude Code's hooks + statusline + stats badge and registers the caveman-shrink MCP proxy on top of the plugin install. Pass --minimal to skip the extras and just install the plugin/extension. Pass --all to also drop per-repo rule files into the current directory.

Flag	What
--all	Plugin + hooks + statusline + MCP shrink + per-repo rule files in $PWD. The full ride.
--minimal	Plugin/extension only. No hooks, no MCP shrink, no per-repo rules.
--dry-run	Preview, write nothing
--only <agent>	One target only (repeatable)
--with-hooks	Claude Code: also wire standalone hooks + statusline + stats badge. On by default.
--with-mcp-shrink	Claude Code: register the caveman-shrink MCP proxy via npx caveman-shrink. On by default.
--with-init	Drop always-on rule files into the current repo (Cursor / Windsurf / Cline / Copilot / AGENTS.md). Off by default; turned on by --all.
--list	Print full agent matrix and exit
--force	Re-run even if already installed

install.sh --help for full reference.

Manual install per agent:

Agent	Command
Claude Code	claude plugin marketplace add JuliusBrussee/caveman && claude plugin install caveman@caveman
Gemini CLI	gemini extensions install https://github.com/JuliusBrussee/caveman
Cursor / Windsurf / Cline / Copilot	npx skills add JuliusBrussee/caveman -a <cursor|windsurf|cline|github-copilot>
Codex / opencode / Roo / Amp / Goose / Kiro / Augment / Aider Desk / Continue / Kilo / Junie / Trae / Warp / Tabnine / Mistral / Qwen / Devin / Droid / ForgeCode / Bob / Crush / iFlow / OpenHands / Qoder / Rovo Dev / Replit / Antigravity	npx skills add JuliusBrussee/caveman -a <profile> (see install.sh --list for the full slug list)
Anything else (40+ agents)	npx skills add JuliusBrussee/caveman (auto-detect)

Standalone Claude Code hooks (without plugin): bash <(curl -s https://raw.githubusercontent.com/JuliusBrussee/caveman/main/hooks/install.sh). Windows: irm https://raw.githubusercontent.com/JuliusBrussee/caveman/main/hooks/install.ps1 | iex. Manual fallback for stubborn Windows envs lives in docs/install-windows.md.

Uninstall: disable the Claude plugin, gemini extensions uninstall caveman, or npx skills remove caveman.

What You Get

Feature	Claude Code	Codex	Gemini CLI	Cursor / Windsurf	Cline / Copilot	Others*
Caveman mode	Y	Y	Y	Y	Y	Y
Auto-activate every session	Y	Y¹	Y	with --with-init	with --with-init	with --with-init
/caveman command	Y	Y¹	Y	—	—	—
Mode switching (lite/full/ultra)	Y	Y¹	Y	Y²	—	—
Statusline badge	Y	—	—	—	—	—
caveman-commit / caveman-review	Y	—	Y	Y	Y	Y
caveman-compress / caveman-help	Y	Y³	Y	Y	Y	Y
caveman-stats	Y	—	—	—	—	—
cavecrew (subagents)	Y	—	—	—	—	—

* opencode, Roo, Amp, Goose, Kiro CLI, Augment, Aider Desk, Continue, Kilo, Junie (JetBrains), Trae, Warp, Tabnine, Mistral, Qwen, Devin, Droid, ForgeCode, Bob, Crush, iFlow, OpenHands, Qoder, Rovo Dev, Replit, Antigravity, and more via npx skills. AGENTS.md / IDE rule files reach Zed, generic agents, etc. via --with-init. ¹ Codex uses $caveman instead of /caveman. Auto-start ships when you run Codex inside this repo (via .codex/hooks.json); for other repos, copy the hook or use $caveman manually. ² Mode switching is on-demand via the skill, no slash command. ³ Compress only.

--with-init writes .cursor/rules/caveman.mdc, .windsurf/rules/caveman.md, .clinerules/caveman.md, .github/copilot-instructions.md, and AGENTS.md into the current repo so caveman auto-starts there.

Usage

Trigger with:

• /caveman or Codex $caveman
• "talk like caveman"
• "caveman mode"
• "less tokens please"

Stop with: "stop caveman" or "normal mode"

Intensity Levels

Level	Trigger	What it do
Lite	/caveman lite	Drop filler, keep grammar. Professional but no fluff
Full	/caveman full	Default caveman. Drop articles, fragments, full grunt
Ultra	/caveman ultra	Maximum compression. Telegraphic. Abbreviate everything

文言文 (Wenyan) Mode

Classical Chinese literary compression — same technical accuracy, but in the most token-efficient written language humans ever invented.

Level	Trigger	What it do
Wenyan-Lite	/caveman wenyan-lite	Semi-classical. Grammar intact, filler gone
Wenyan-Full	/caveman wenyan	Full 文言文. Maximum classical terseness
Wenyan-Ultra	/caveman wenyan-ultra	Extreme. Ancient scholar on a budget

Level stick until you change it or session end.

Caveman Skills

Skill	What
/caveman-commit	Terse commit messages. Conventional Commits, ≤50 char subject. Why over what.
/caveman-review	One-line PR comments: L42: 🔴 bug: user null. Add guard. No throat-clearing.
/caveman-help	Quick-reference card. All modes, skills, commands.
/caveman-stats	Real session token usage + estimated savings + USD. Lifetime aggregation via --all, time window via --since 7d, tweetable line via --share. Reads the Claude Code session JSONL directly, no model-side guessing. Claude Code only.
/caveman:compress <file>	Rewrites a memory file (e.g. CLAUDE.md) into caveman-speak. Saves backup as <file>.original.md. Cuts ~46% of input tokens every session start. Code/URLs/paths preserved byte-for-byte.
cavecrew-investigator/builder/reviewer	Caveman subagents for Claude Code. Subagent tool-output gets injected back into main context — these emit ~60% fewer tokens than vanilla Explore / reviewer agents, so main context lasts longer across long sessions. Investigator (read-only locator, haiku), builder (1-2 file surgical edit, refuses 3+), reviewer (one-line findings, haiku).

Statusline savings badge — on by default. After your first /caveman-stats run the statusline appends [CAVEMAN] ⛏ 12.4k (lifetime tokens saved) and updates every time /caveman-stats runs. Don't want it? Set CAVEMAN_STATUSLINE_SAVINGS=0 to silence.

caveman-compress receipts

File	Original	Compressed	Saved
claude-md-preferences.md	706	285	59.6%
project-notes.md	1145	535	53.3%
claude-md-project.md	1122	636	43.3%
todo-list.md	627	388	38.1%
mixed-with-code.md	888	560	36.9%
Average	898	481	46%

Full docs: caveman-compress README. Snyk false-positive note.

caveman-shrink (MCP middleware)

Stdio proxy that wraps any MCP server, intercepts tools/list / prompts/list / resources/list responses, and compresses the description fields. Code, URLs, paths, identifiers stay byte-for-byte identical.

{
  "mcpServers": {
    "fs-shrunk": {
      "command": "npx",
      "args": ["caveman-shrink", "npx", "@modelcontextprotocol/server-filesystem", "/path/to/dir"]
    }
  }
}

Published on npm as caveman-shrink. V1 does not touch tool-call response bodies or request payloads. Auto-registered by install.sh (use --minimal to skip). Full docs: mcp-servers/caveman-shrink/.

Benchmarks

Real token counts from the Claude API (reproduce it yourself):

Task	Normal (tokens)	Caveman (tokens)	Saved
Explain React re-render bug	1180	159	87%
Fix auth middleware token expiry	704	121	83%
Set up PostgreSQL connection pool	2347	380	84%
Explain git rebase vs merge	702	292	58%
Refactor callback to async/await	387	301	22%
Architecture: microservices vs monolith	446	310	30%
Review PR for security issues	678	398	41%
Docker multi-stage build	1042	290	72%
Debug PostgreSQL race condition	1200	232	81%
Implement React error boundary	3454	456	87%
Average	1214	294	65%

Range: 22%–87% savings across prompts.

[!IMPORTANT] Caveman only affects output tokens — thinking/reasoning tokens are untouched. Caveman no make brain smaller. Caveman make mouth smaller. Biggest win is readability and speed, cost savings are a bonus.

A March 2026 paper "Brevity Constraints Reverse Performance Hierarchies in Language Models" found that constraining large models to brief responses improved accuracy by 26 percentage points on certain benchmarks and completely reversed performance hierarchies. Verbose not always better. Sometimes less word = more correct.

Evals

Caveman not just claim 75%. Caveman prove it.

The evals/ directory has a three-arm eval harness that measures real token compression against a proper control — not just "verbose vs skill" but "terse vs skill". Because comparing caveman to verbose Claude conflate the skill with generic terseness. That cheating. Caveman not cheat.

# Run the eval (needs claude CLI)
uv run python evals/llm_run.py

# Read results (no API key, runs offline)
uv run --with tiktoken python evals/measure.py

Star This Repo

If caveman save you mass token, mass money — leave mass star. ⭐

🪨 The Caveman Ecosystem

Three tools. One philosophy: agent do more with less.

Repo	What	One-liner
caveman (you are here)	Output compression skill	why use many token when few do trick — ~75% fewer output tokens across Claude Code, Cursor, Gemini, Codex
cavemem	Cross-agent persistent memory	why agent forget when agent can remember — compressed SQLite + MCP, local by default
cavekit	Spec-driven autonomous build loop	why agent guess when agent can know — natural language → kits → parallel build → verified

They compose: cavekit orchestrates the build, caveman compresses what the agent says, cavemem compresses what the agent remembers. Install one, some, or all — each stands alone.

Also by Julius Brussee

• Revu — local-first macOS study app with FSRS spaced repetition, decks, exams, and study guides. revu.cards

License

MIT — free like mass mammoth on open plain.