skills.rest
Antony-Figueroa
Antony-FigueroaCOMMUNITY
Explore all skills by this author →
GitHub

caveman

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

3|1|Updated Dec 25, 2025
Search Tags:#persistence#token-efficiency#caveman#ai-safety#intensity-levels#output-compression
One-click install
npx skills add https://github.com/Antony-Figueroa/UNEFA_DASHBOARD --skill caveman-antony-figueroa
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: caveman
Source: https://github.com/Antony-Figueroa/UNEFA_DASHBOARD/tree/main/.trae/skills/caveman
Command: npx skills add https://github.com/Antony-Figueroa/UNEFA_DASHBOARD --skill caveman-antony-figueroa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Caveman mode compresses model responses into concise caveman-style prose while preserving all technical content, including code blocks, error strings, and symbols.

Core Features & Use Cases

  • Ultra-compressed output across responses with six intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra.
  • Persistence across a session and auto-clarity rules to avoid removing critical safety checks, ensuring reliable and safe interactions.
  • Clear activation and boundaries: supports commands like /caveman and "caveman mode" to switch modes, with a fast resume to normal mode when requested.

Quick Start

Invoke caveman mode by sending '/caveman' or 'caveman mode' to start 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 chat responses for software engineering tasks?▼

To reduce token usage in AI chat responses, you can compress model outputs into concise caveman-style prose while preserving all technical content, including code blocks and error strings. This approach maintains technical accuracy across long-running interactive sessions.

What is caveman-style output compression for AI sessions?▼

Caveman-style output compression is a technique that transforms model responses into terse, high-precision prose while retaining all technical accuracy. It operates across six intensity levels and includes auto-clarity rules to ensure safety checks are not removed during interactive discussions.

How do I activate caveman mode to get terse AI responses?▼

You activate caveman mode by sending the command '/caveman' or the phrase 'caveman mode' to start receiving terse, high-precision responses. The mode persists across the session until you request a fast resume back to normal output.

Does output compression remove critical safety checks from code explanations?▼

Output compression does not remove critical safety checks due to built-in auto-clarity rules and boundary safeguards. The system ensures predictable behavior and safety by preserving all technical content, including error strings and symbols, even at ultra intensity levels.

Can I adjust the intensity level of output compression for technical queries?▼

You can adjust the intensity level of output compression using six available settings: lite, full, ultra, wenyan-lite, wenyan-full, and wenyan-ultra. The default intensity is set to full, allowing you to scale the terseness based on your specific needs.

Related Skills

Explore More Software Engineering Skills→
harukainguyen1411harukainguyen1411

pre-compact-save

Consolidate session memory and state before context compression.

Community
Intermediate
AlbertLin821AlbertLin821
2

caveman

Compress model responses into terse caveman-style prose with intensity levels.

Community
Intermediate
vanyastaffvanyastaff
2

ctx-stats

Display token consumption, context savings ratio, and per-tool breakdown for session context-mode usage.

Community
Basic
samhvw8samhvw8
14

prompt-enhancer

Transform verbose prompts into concise, XML-structured versions for multiple AI models.

Community
Basic