caveman

Compresses technical communication to reduce token usage while preserving accuracy.

Updated Apr 23, 2026

Search Tags:#token-optimization #prompting #caveman #ai-communication #intensity-levels #tech-accuracy #terse-output

Ready to run

npx skills add https://github.com/giovannimnz/gsd-caveman-hermes --skill caveman-giovannimnz

💬 Prompt for ChatGPT / Claude Chat▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Compresses technical communication to reduce token usage while preserving technical accuracy.

Core Features & Use Cases

Ultra-compact responses with adjustable intensity levels: lite, full, ultra, and wenyan variants.
Persistence across turns and auto-activation when token efficiency is requested.
Clear, terse explanations without fluff, suitable for debugging and design discussions.

Quick Start

Enable caveman mode by saying '/caveman' or 'caveman mode' to activate terse, high-precision 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 AI communication while keeping technical accuracy?▼

To reduce token usage while maintaining technical accuracy, enable compressed technical communication. This Skill compresses development discussions and debugging explanations into terse, high-precision outputs without losing necessary technical details.

What is terse technical output and when should I use it for debugging?▼

Terse technical output is ultra-compact communication that removes conversational fluff while preserving technical accuracy. It is ideal for debugging sessions and design reviews where context is established and concise responses are preferred.

How do I activate caveman mode for prompt optimization?▼

Activate prompt optimization by typing '/caveman' or 'caveman mode' in your prompt. This enables terse, high-precision responses that persist across turns to continuously reduce token usage.

Can I adjust the intensity level of token optimization during a conversation?▼

Yes, token optimization intensity is adjustable across lite, full, ultra, and wenyan variants. The mode persists across turns, defaulting to full mode, allowing you to change compression levels as needed.

Does token-optimized communication work for code explanations and design reviews?▼

Yes, token-optimized communication applies directly to code explanations and design reviews. It provides clear, terse explanations without fluff, ensuring technical accuracy is preserved while minimizing token consumption across development discussions.