skills.rest
t-jet
t-jetCOMMUNITY
Explore all skills by this author →
GitHub

caveman

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

Updated Aug 11, 2026
Search Tags:#token efficiency#brevity#token compression#classical chinese#response style#caveman mode#prompt behavior
One-click install
npx skills add https://github.com/t-jet/pal_found_cli --skill caveman-t-jet
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: caveman
Source: https://github.com/t-jet/pal_found_cli/tree/main/.ept/skills/caveman
Command: npx skills add https://github.com/t-jet/pal_found_cli --skill caveman-t-jet

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Long, filler-heavy AI responses waste tokens and slow down reading. This Skill cuts output token usage by roughly 65-75% by rewriting every response in a compressed caveman register, dropping articles, filler, and hedging while keeping code, error strings, and technical terms exact. ## Core Features & Use Cases - Six intensity levels: lite, full (default), ultra, plus three Classical Chinese (文言文) variants for maximum character reduction. - Session persistence: the mode stays active across all turns until explicitly stopped with "stop caveman" or "normal mode". - Auto-clarity fallback: automatically reverts to normal prose for security warnings, irreversible-action confirmations, and ambiguous multi-step sequences, then resumes compression. - Use Case: During a long debugging session, invoke /caveman ultra so every explanation arrives as terse fragments like "Inline obj prop → new ref → re-render. useMemo.", cutting reading time without losing precision. ## Quick Start Ask the assistant to enable caveman mode by saying "use caveman mode" or invoking /caveman, and all subsequent responses will be compressed until you say "stop caveman".

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 by saying "caveman mode", "be brief", or invoking /caveman. The assistant then drops articles, filler, and hedging from every response, cutting output tokens by roughly 65-75% while keeping code and error strings exact.

How do I turn off caveman mode once it is active?▼

Say "stop caveman" or "normal mode" and the assistant immediately reverts to standard prose. The mode otherwise persists for the entire session across all turns.

Does compressed output lose technical accuracy or code details?▼

No. Code blocks, function names, API names, and error strings are never abbreviated or altered. Only prose elements like articles, pleasantries, and hedging are removed, so technical substance stays intact.

What are the differences between lite, full, and ultra compression levels?▼

Lite removes filler but keeps full sentences; full drops articles and allows fragments; ultra adds abbreviations like DB and auth plus arrows for causality. Three Classical Chinese levels (wenyan-lite/full/ultra) offer up to 80-90% character reduction.

When does caveman mode automatically switch back to normal prose?▼

It auto-reverts for security warnings, irreversible-action confirmations, multi-step sequences where fragments risk misreading, and when you repeat a question. Normal prose resumes only for that clear portion, then compression continues.

Related Skills

Explore More Content & Communication Skills→
RDCrystalLabRDCrystalLab

caveman

Compress AI responses into terse modes while preserving technical accuracy.

Official
Basic
valkryhxvalkryhx
6

opencode

Execute autonomous coding tasks with multi-file refactoring and iterative debugging.

Community
Advanced
mscipiomscipio

token-tracker

Monitor AI model context window usage and warn before token overflow.

Community
Advanced
JNRCHAYANJNRCHAYAN

caveman

Compresses assistant responses into terse, technically accurate language.

Community
Intermediate