caveman

Compress language model output into caveman-style prose with six intensity levels.

Updated May 25, 2026

Search Tags:#communication #compression #text compression #caveman style #response reduction

Ready to run

npx skills add https://github.com/ngquaanghuy/Crypto --skill caveman-ngquaanghuy

💬 Prompt for ChatGPT / Claude Chat▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps in compressing model responses into a caveman-style prose, reducing the output token count while preserving technical accuracy.

Core Features & Use Cases

Response Compression: Compresses model responses by dropping unnecessary words and phrases while maintaining technical details.
Intensity Levels: Offers six intensity levels for varying compression levels, from minimal to extreme.
Auto-Clarity: Switches to normal prose for security warnings, irreversible actions, and when clarity is needed.
Language Preservation: Preserves the user's dominant language for consistent responses.
Usage Scenarios: Useful for users who want to quickly understand the essence of a response without overwhelming details.

Quick Start

Start the caveman mode and get compressed responses.

Frequently Asked Questions about caveman

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

FAQPage Schema

How do I compress language model output to reduce token usage?▼

Compress language model output by converting responses to caveman-style prose, dropping unnecessary words while preserving technical accuracy. This reduces token usage by 75% and is useful for quickly understanding the essence of a response.

Can I control the intensity of text compression for different scenarios?▼

Control text compression using six intensity levels, ranging from minimal to extreme. This allows you to adjust the brevity of model responses based on specific communication needs, such as technical support or documentation review.

Does response compression switch to normal prose for critical information?▼

Response compression auto-switches to normal prose for critical information, including security warnings and irreversible actions. This auto-clarity feature ensures that essential technical details remain clear and unambiguous.

Will text compression preserve the dominant language of my prompts?▼

Text compression preserves the user's dominant language to ensure consistent responses. This language preservation feature maintains communication continuity while applying the desired level of response reduction.

What is caveman-style prose and when do I need it for technical support?▼

Caveman-style prose is a brief, stripped-down communication format that removes filler words while retaining technical details. It is needed when you want to quickly review documentation or technical support responses without overwhelming details.

Are there limitations to using response reduction for technical documentation?▼

Response reduction is limited by its auto-clarity feature, which overrides extreme compression for security warnings and irreversible actions. It should not be used when full, verbose documentation context is explicitly required.