caveman

Compresses AI output to minimize token usage while preserving technical accuracy.

Updated Jan 12, 2026

Search Tags:#efficiency #compression #token-optimization #technical-writing #terse

Ready to run

npx skills add https://github.com/giosuetedeschi-spec/bobu-website --skill caveman-giosuetedeschi-spec

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: caveman
Source: https://github.com/giosuetedeschi-spec/bobu-website/tree/main/.claude/skills/caveman
Command: npx skills add https://github.com/giosuetedeschi-spec/bobu-website --skill caveman-giosuetedeschi-spec

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses high token consumption by enforcing a hyper-terse, compressed communication style that strips away filler and fluff while maintaining technical precision.

Core Features & Use Cases

Token Compression: Reduces output length by up to 65% by removing articles, pleasantries, and hedging.
Intensity Control: Offers multiple levels of compression ranging from professional-lite to extreme classical Chinese (wenyan) styles.
Use Case: Ideal for long-running technical sessions where maintaining context window space is critical for complex debugging or architectural discussions.

Quick Start

Activate the caveman skill by typing caveman mode to immediately begin receiving ultra-compressed technical responses.

Frequently Asked Questions about caveman

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I reduce token usage in long-running AI chat sessions?▼

To reduce token usage, you can compress AI output into a terse, high-density format that strips away filler while maintaining technical accuracy. This preserves context window space during complex debugging or architectural discussions.

What is ultra-compressed technical writing for LLMs?▼

Ultra-compressed technical writing is a communication style that removes articles, pleasantries, and hedging to reduce output length by up to 65%. It maintains technical verbatim integrity for code, CLI commands, and error strings.

Can I control the intensity of token compression for AI responses?▼

Yes, you can control token compression intensity through multiple levels ranging from standard professional-lite to extreme classical Chinese modes. This accommodates diverse linguistic requirements while optimizing token density.

Does token compression preserve code and CLI command accuracy?▼

Token compression preserves technical verbatim integrity for code, CLI commands, and error strings. It strips away natural language fluff while ensuring that exact technical commands remain untouched and fully functional.

When should I use terse communication modes for AI output?▼

Use terse communication modes during long-running technical sessions where context window space is critical. It is ideal for complex debugging or architectural discussions that require sustained context over many interactions.

What are the limitations of hyper-terse AI output formats?▼

Hyper-terse AI output formats strip away pleasantries and hedging, which may reduce readability for non-technical contexts. They are highly specialized for token optimization and technical precision rather than conversational or explanatory use.