skills.rest
alexeyshishin
alexeyshishinCOMMUNITY
Explore all skills by this author →
GitHub

caveman

Compresses AI responses into terse fragment-style output with six intensity levels.

4|Updated Aug 8, 2026
Search Tags:#communication style#token efficiency#brevity#wenyan#caveman mode#compressed output#prompt behavior
One-click install
npx skills add https://github.com/alexeyshishin/as-skill --skill caveman-alexeyshishin
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: caveman
Source: https://github.com/alexeyshishin/as-skill/tree/main/domains/core/skills/caveman
Command: npx skills add https://github.com/alexeyshishin/as-skill --skill caveman-alexeyshishin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Long, filler-heavy AI responses waste output tokens and slow down reading when you only need the technical answer. This Skill enforces an ultra-compressed communication style that cuts output tokens by roughly 65% while preserving full technical accuracy. ## Core Features & Use Cases - Six intensity levels: lite, full (default), ultra, plus three Classical Chinese (wenyan) variants for maximum character reduction. - Persistent mode: stays active across the whole session until explicitly disabled with "stop caveman" or "normal mode". - Auto-clarity fallback: automatically drops compression for security warnings, irreversible action confirmations, and ambiguous multi-step instructions. - Use Case: During a long debugging session, say "caveman mode" and receive terse, direct answers like "Bug in auth middleware. Token expiry check use < not <=." instead of paragraph-long explanations. ## Quick Start Tell the assistant "use caveman mode" or invoke /caveman to switch all subsequent responses to compressed output.

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?▼

Activate caveman mode by saying "caveman mode", "be brief", or invoking /caveman. The assistant then drops articles, filler words, and pleasantries while keeping all technical content exact, cutting output tokens by about 65%.

What intensity levels does caveman mode support?▼

Six levels: lite (no filler, full sentences), full (default, drops articles and uses fragments), ultra (maximum English compression), and wenyan-lite, wenyan-full, wenyan-ultra for Classical Chinese terseness. Switch with /caveman followed by the level name.

Does caveman mode work in languages other than English?▼

Yes, it preserves the user's dominant language exactly and only compresses style. Article-dropping applies only to article languages; grammatical particles in other languages are kept, and dedicated wenyan modes handle Classical Chinese.

When does caveman mode turn off automatically?▼

It temporarily drops compression for security warnings, irreversible action confirmations, and multi-step sequences where fragments could cause misreading, then resumes afterward. It fully stops only on "stop caveman" or "normal mode".

Does compressed style affect code, commits, or documents the AI writes?▼

No. Code blocks, error strings, technical terms, and API names stay verbatim in chat, and anything persisted outside the conversation such as code, comments, commits, docs, and PR text is written in normal prose.

Related Skills

Explore More Personal & Entrepreneur Skills→
MauricioDMOMauricioDMO

caveman

Compresses AI responses into terse fragments while preserving technical accuracy.

Community
Basic
constellizeconstellize
1

revise-memory-bank

Document structured revisions across memory-bank change-control artifacts.

Official
Advanced
ProgramadorBrasilProgramadorBrasil

context-agent

Capture, compress, and restore conversation context across Claude Code sessions.

Community
Advanced
dhar174dhar174
1

create-github-action-workflow-specification

Generate implementation-agnostic specification documents from GitHub Actions workflow files.

Community
Intermediate