Save money by shortening AI responses
Caveman is a capability package developed for Claude Code that reduces token consumption by 65 percent by simplifying the language use of the artificial intelligence model. This tool aims to reduce processing costs by reducing complex commands to a primitive language structure.
Updates
- November 8, 2026: Stars 117,026 → 118,195, latest release v4.1.0 (November 8, 2026).
- November 3, 2026: Stars 111,432 → 117,026, latest release v4.0.1 (October 13, 2026).
- October 13, 2026: Stars 110,510 → 111,432, latest release v4.0.1 (October 13, 2026).
- October 8, 2026: Stars 109,692 → 110,510, latest release v3.2.0 (October 8, 2026).
What you get
- Saves up to 65% tokens on AI responses.
- Eliminates unnecessary words without sacrificing technical accuracy.
- It reduces transaction costs and increases response speed.
Installation
curl -fsSL https://raw.githubusercontent.com/JuliusBrussee/caveman/main/install.sh | bashirm https://raw.githubusercontent.com/JuliusBrussee/caveman/main/install.ps1 | iexRunning it
/cavemanIf you don't write code
From now on, give your answers in Caveman mode. Remove unnecessary filler words, keep only the concise information and technical details. Compose your answers in short, concise and fragmented sentences; but continue to convey complete and precise code blocks, commands and error messages.
Repository: https://github.com/JuliusBrussee/caveman. Follow the official instructions in the README of this repository for setup and execution.
Before starting, check which OS we are on and if the necessary tools (e.g., Python, Node.js, Git, Docker) are installed. Install any missing ones step-by-step according to this system.
Create a new, empty folder to work in. Do not touch any files outside this folder and do not delete anything. Tell me what you are going to do in one sentence before running each command.
Tell me what I should see on the screen at the end of each step. If an error occurs, I will paste the error message to you exactly as it is, and we will fix it together.
Related dictionary terms
Links
TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-07-03: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. The introductory text was prepared with AI from the repository's own documentation, and this page was machine-translated from the Turkish original · the Turkish version prevails.