skills.rest
filippolmt
filippolmtCOMMUNITY
Explore all skills by this author →
GitHub

caveman

Compresses AI responses into terse caveman-style prose while preserving technical accuracy.

1|Updated Jul 2, 2026
Search Tags:#brevity#token compression#terse output#wenyan#response style#caveman mode#prompt behavior
One-click install
npx skills add https://github.com/filippolmt/skills --skill caveman-filippolmt
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: caveman
Source: https://github.com/filippolmt/skills/tree/main/skills/caveman
Command: npx skills add https://github.com/filippolmt/skills --skill caveman-filippolmt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Long, filler-heavy AI responses waste tokens and slow down reading. This Skill switches the assistant into an ultra-compressed communication mode that drops articles, filler, pleasantries, and hedging while keeping code, error strings, and technical terms exact. ## Core Features & Use Cases - Six intensity levels: lite, full, ultra, plus three wenyan (Classical Chinese) variants for maximum character compression. - Auto-clarity fallback: Automatically reverts to normal prose for security warnings, irreversible-action confirmations, and ambiguous multi-step sequences, then resumes compression. - Session persistence: The chosen style applies to every response until the user says "stop caveman" or changes the level. - Use Case: During a long debugging session, enable full mode so every explanation arrives as terse fragments like "Bug in auth middleware. Token expiry check use < not <=.", cutting reading time without losing the fix. ## Quick Start Ask the assistant to talk like caveman or enable caveman mode to make all subsequent responses terse and token-efficient.

Frequently Asked Questions about caveman

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

FAQPage Schema
How do I make AI responses shorter and use fewer tokens?▼

Enable caveman mode, which drops articles, filler words, pleasantries, and hedging from every response. Technical terms, code blocks, and error strings stay exact, so answers stay accurate while becoming much shorter.

How to switch between caveman compression levels?▼

Invoke the skill with a level argument: lite keeps full sentences, full drops articles and allows fragments, ultra strips down to bare facts, and the wenyan variants produce Classical Chinese. The level persists until changed or the session ends.

Does compressed output stay accurate for code and errors?▼

Yes. Code blocks, function names, API names, and error strings are never altered, and negations like not/never/only are always kept. Only stylistic filler is removed, never technical substance.

When does caveman mode turn itself off automatically?▼

It drops to normal prose for security warnings, irreversible-action confirmations, multi-step sequences where fragments could be misread, and when the user repeats a question. Compression resumes after the clear part is done.

Does caveman mode affect code, commits, and documentation I write?▼

No. The compressed style applies only to chat responses. Code, comments, commit messages, docs, issues, and messages to other people are always written in normal prose.

Related Skills

Explore More Content & Communication Skills→
YassimbaYassimba
9

wtf

Restates the last message in plain language with diagrams and from-scratch explanations.

Community
Intermediate
itallstartedwithaideaitallstartedwithaidea
31

token-optimization

Optimizes token usage via routing, compression, caching, async processing, and budgets for AI agent workloads.

Community
Advanced
dallasbpetersdallasbpeters

shape-your-agent

Craft a tailored system prompt for Sanity Agent Context MCP agents.

Community
Advanced
ronniegeraghtyronniegeraghty
3

economy-mode

Reduce Layer 3 model selection to lower-cost alternatives when active.

Community
Intermediate