caveman

Compress AI responses into caveman-style prose while preserving code and error text.

Updated Aug 23, 2026

Search Tags:#technical writing #token efficiency #prompting #classical chinese #response compression #style control

Ready to run

npx skills add https://github.com/ebarroso12/personal-skills --skill caveman-ebarroso12

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: caveman
Source: https://github.com/ebarroso12/personal-skills/tree/main/caveman
Command: npx skills add https://github.com/ebarroso12/personal-skills --skill caveman-ebarroso12

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reduces the verbosity and token usage of AI responses while preserving technical accuracy, so conversations stay efficient without losing critical details like code blocks, error strings, and exact symbols.

Core Features & Use Cases

Ultra-compressed output mode: Speaks in caveman-style prose by dropping articles, filler, hedging, and pleasantries while keeping technical substance intact.
Intensity levels: Switch between lite, full (default), ultra, and multiple Wenyan (classical Chinese) registers for different compression needs.
Safety auto-clarity: Automatically returns to normal prose for security warnings, irreversible confirmations, ambiguity-prone multi-step sequences, and when the user repeats or asks to clarify.

Quick Start

Ask your AI to respond in caveman mode by saying: /caveman ultra.

Frequently Asked Questions about caveman

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

FAQPage Schema

How do I reduce AI response token usage without losing technical accuracy?▼

To reduce AI response token usage without losing technical accuracy, use a compression Skill that drops filler and hedging while keeping code blocks, error strings, and exact symbols unchanged.

How do I switch AI response compression intensity levels?▼

Switch AI response compression intensity levels by issuing a command like /caveman ultra, allowing you to toggle between lite, full, ultra, and classical Chinese Wenyan registers for different needs.

Does response compression automatically switch back to normal prose for security warnings?▼

Yes, response compression automatically switches back to normal prose for security warnings, irreversible confirmations, ambiguity-prone multi-step sequences, and when the user asks to clarify.

What is the best way to compress AI chat outputs for token efficiency?▼

The best way to compress AI chat outputs for token efficiency is applying a caveman-style prose mode that strips articles and pleasantries while preserving exact structured details and technical substance.

When should I not use caveman-style response compression?▼

You should not use caveman-style response compression when handling security warnings, irreversible actions, or ambiguous multi-step sequences, as the Skill auto-reverts to normal prose to ensure clarity.