caveman

🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman

244 articles 82.8k View on GitHub ↗
244 articles
How Caveman Achieves Token Savings on Agent Outputs and Inputs: 2 Core Mechanisms Explained

Discover how Caveman slashes token costs with tool schema reduction and context compaction. Learn its 2 core mechanisms for saving on agent inputs and outputs.

deep-dive
Sep 6, 2026
Where Are the Caveman Compressor Implementations Located?

Discover where Caveman compressor implementations are located in the JuliusBrussee/caveman repository. Find dedicated Go files for JSON text HTML code diff log and more within the engine compressors package.

how-to-guide
Sep 6, 2026
How to Check Session Token Usage Statistics with Caveman: A Complete Guide

Easily check session token usage statistics with Caveman. Learn how the built-in caveman-stats hook displays detailed token numbers and estimated savings for your Claude Code sessions.

how-to-guide
Sep 6, 2026
Complete List of Caveman Slash Commands in Claude Code

Discover all Caveman slash commands for Claude Code. Explore file operations, compression, and utility functions invoked with a leading /.

api-reference
Sep 6, 2026
How Caveman Mode State is Managed Per Session in Claude Code

Discover how Caveman mode state is managed per session in Claude Code. Learn about its session-specific lifecycle and lack of cross-invocation persistence for efficient agent runs.

internals
Sep 6, 2026
How Claude Code Hooks Integrate with Caveman: Complete Technical Guide

Discover how Claude Code hooks integrate with Caveman. This guide details SessionStart, UserPromptSubmit, and Statusline hooks for mode initialization, prompt tracking, and visual feedback.

how-to-guide
Sep 6, 2026
Caveman Skill Intensity Levels: The Complete Guide (6 Modes Explained)

Explore Caveman skill intensity levels: lite, full, ultra, wenyan-lite, wenyan-full, and wenyan-ultra. Control prose compression with this complete guide.

deep-dive
Sep 6, 2026
Caveman Skills Explained: Structure, Configuration, and Usage in the Caveman LLM Proxy

Learn how to structure, configure, and use Caveman skills. These modular plug-ins transform LLM responses and expose new commands via markdown files with YAML front-matter and plain-text rules.

deep-dive
Sep 6, 2026
Where to Find Caveman Agent Profile Definitions: Location and Usage Guide

Discover where to find Caveman agent profile definitions. Locate and understand these JSON files in the agents profiles directory for your AI agent configurations.

how-to-guide
Sep 6, 2026
Caveman Wrap Command-Line Options: How to Configure Agent Sessions with Flags

Learn Caveman wrap command-line options to configure agent sessions with flags like --off --pixel --workflow and --full-auto. Control your agent's behavior efficiently.

how-to-guide
Sep 6, 2026
How Caveman Wraps Existing Coding Agents like Claude Code and Codex: A Technical Deep-Dive

Discover how Caveman wraps coding agents like Claude Code and Codex. Learn about its innovative proxy injection technique for seamless integration without code modification.

deep-dive
Sep 6, 2026
What Are the Different Proxy Modes Available in Caveman?

Explore Caveman's three proxy modes: default, MCP-shrink, and legacy HTTP proxy. Learn how to optimize your caching with these flexible options.

how-to-guide
Sep 6, 2026
…

Have a question about this repo?

These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:

Share the following with your agent to get started:
curl -s "https://instagit.com/install.md"
Works with
Claude Codex Cursor VS Code OpenClaw Any MCP Client

Maintain an open-source project? Get it listed too →