caveman

AI & Agents
Julius Brussee
Your coding agent leads with the answer and drops the filler, while code and error messages stay character for character.
REPO
JuliusBrussee/caveman
INSTALL
npx skills add, or a Claude Code plugin
NEEDS
Claude Code and 30+ other agents
Last updated
November 13, 2026

Preview

caveman

What it is.

caveman is a Claude skill that changes how your coding agent replies. Each answer comes first, with no greeting or recap, and code and error messages stay exactly as written. It stays on for the whole session until you say stop caveman.

That matters because every word the agent writes is billed as an output token, and you read it before you act. A paired test by JetBrains across 86 coding tasks found caveman cut output tokens by 8.5%, with no measurable loss in quality.

It switches back to full sentences for security warnings and irreversible actions, and anything saved outside the chat, such as commits and docs, stays in normal prose. The skill is one part of a larger toolkit, whose full install also adds a proxy and model routing.

What you get.

  • Replies that lead with the answer and end with the next step.
  • Code blocks, commands, file paths and error messages left exactly as written.
  • Full sentences for security warnings and before any irreversible action.
  • Normal prose in anything saved outside the chat, such as commits and docs.
  • Two stricter modes, /ultracave and /megacave, for even shorter replies.
  • A one-line way out: say stop caveman or normal mode.
HOW TO USE IT

How to set it up.

1

Install the skill on its own with npx skills add JuliusBrussee/caveman -g.

2

Or, inside Claude Code, run /plugin marketplace add JuliusBrussee/caveman, then /plugin install caveman@caveman.

3

Type /caveman if it doesn't start on its own, or $caveman in Codex.

4

Use /ultracave when you want the shortest replies on routine work.

5

Say stop caveman or normal mode to go back to normal replies.

6

Leave out the full npx @caveman-ai/cli install unless you also want its proxy and model routing.

Pricing Plans

Free and open source under Apache-2.0.

Use cases

Long Claude Code sessions

You spend an afternoon building a lead-scoring pipeline in Claude Code. Replies stay short, so you scroll less and pay for fewer output tokens.

Debugging a broken workflow

You paste an n8n error into Claude Code. The reply names the cause and the change, with the error quoted exactly.

Working in a small terminal

Claude Code running over SSH on a laptop screen leaves little room. Short replies put the command near the top.

Best for

GTM engineers

They build workflows and scripts in Claude Code and want the agent's replies out of the way.

RevOps leads who script

They read agent output in a terminal and care about the exact command more than the explanation.

Teams paying per token

Shorter replies cost less when the bill counts output tokens.

Read the source

Published by Julius Brussee. Opens in a new tab.
Open the Tool

Questions about caveman

Does caveman make Claude's answers worse?
How much does it save?
Will it change what I type?
Will my commits and docs come out in caveman?

Questions about AI & Agents

What is a Claude skill?
What are Claude coding skills?
Do shorter-answer skills save money?
Strengths
  • Code and error messages are kept character for character.
  • Switches back to full sentences for security warnings and irreversible actions.
  • Publishes the cases where it loses, including per-request billing.
  • Installs on its own, apart from the proxy and model routing.
Limitations
  • Against a plain answer concisely instruction, the default mode saves only 3% more output tokens at the median.
  • The rules add about 1,160 input tokens, and on per-request billing such as GitHub Copilot, shorter answers save nothing.
  • The Claude app install path is marked as not tested end to end yet.
  • The full npx install changes more than replies, so read the list of changes it shows before you confirm.
Skip this if
  • Skip it if your plan bills per request and you already ask Claude to answer concisely.
Ideal for
  • GTM engineers
  • RevOps leads who script
  • Teams paying per token

The team behind these plays.

We build inbound GTM engines for B2B software teams, and these are the plays we build from. Tell us the pipeline target and we'll show the plan under it.