# Caveman

> Caveman makes coding agents say less and read less: terse answers, logs and JSON shrunk before the model reads them, compressed web pages, reusable scripts in code mode, and optional Auto model routing. The README reports 33.2% fewer input tokens across 54 Claude Code runs. One npx install (Node.js 22.13+) on Mac, Linux or Windows. Apache-2.0. The CLI sends usage stats by default (no prompts or code); Auto needs a free account and sends your asks to Caveman Cloud.

- Repository: https://github.com/JuliusBrussee/caveman
- Official docs: https://github.com/JuliusBrussee/caveman#readme
- Checked: 2026-10-23

## What it replaces
Nothing people pay for: a free tool, listed because people search for it.

## Facts, with sources

| Question | Status | Answer | Source | Checked |
|---|---|---|---|---|
| Runs on my machine? | Confirmed | Yes, the skill and CLI run on your machine with your own model keys or login | https://github.com/JuliusBrussee/caveman#install | 2026-10-23 |
| Code licence | Confirmed | Apache-2.0, whole repo | https://github.com/JuliusBrussee/caveman#license | 2026-10-23 |
| Model weights licence | Confirmed | No bundled weights; uses your coding agent's model | https://github.com/JuliusBrussee/caveman#install | 2026-10-23 |
| Commercial use | Confirmed | Yes, Apache-2.0; "Caveman" and the logo are trademarks of the author | https://github.com/JuliusBrussee/caveman#license | 2026-10-23 |
| Paid services needed? | Partial | Needs a coding agent and its model (a paid plan or API key); Auto routing needs a free Caveman account | https://github.com/JuliusBrussee/caveman#auto-54-cheaper-same-score | 2026-10-23 |
| Hardware | Confirmed | No hardware needs of its own; needs Node.js 22.13+ for the CLI | https://github.com/JuliusBrussee/caveman#install | 2026-10-23 |
| Operating systems | Confirmed | macOS, Linux and Windows | https://github.com/JuliusBrussee/caveman#install | 2026-10-23 |
| Setup effort | Confirmed | Easy, one command (npx @caveman-ai/cli@latest) that finds your agents and asks before changing them | https://github.com/JuliusBrussee/caveman#install | 2026-10-23 |

GitHub: 116945 stars, 6795 forks, last push 2026-11-02, archived: no.

## Alternatives to Caveman
- [Superpowers](https://repoquarry.dev/projects/superpowers.md): Skill library for coding agents that walks them through brainstorming, planning, test-driven development, subagent work and code review.
- [Skills For Real Engineers (Matt Pocock)](https://repoquarry.dev/projects/mattpocock-skills.md): Matt Pocock's small, composable agent skills for engineering work, such as grilling sessions, TDD, bug diagnosis, specs and code review.
- [ECC](https://repoquarry.dev/projects/ecc.md): Large plugin of agents, skills, hooks, rules and memory for Claude Code, with native Codex and Qoder plugins and partial adapters for other agents.
- [Ponytail](https://repoquarry.dev/projects/ponytail.md): A skill (one prompt) for AI coding agents that makes them reuse what exists and write less code, plus /ponytail-review and /ponytail-audit commands.
- [UI UX Pro Max](https://repoquarry.dev/projects/ui-ux-pro-max.md): Design skill for Claude Code, Cursor, Codex and other agents that picks UI styles, palettes, fonts and UX rules and generates a design system.
- [Graphify](https://repoquarry.dev/projects/graphify.md): A /graphify skill that maps a codebase and its docs into a knowledge graph your coding agent queries instead of grepping files.
---
Source: RepoQuarry (https://repoquarry.dev/projects/caveman). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.