caveman

Compress AI communication by removing filler and hedging to reduce token usage.

Updated May 11, 2026

Search Tags:#debugging #technical writing #token reduction #caveman mode #prompt formatting #terse responses

Ready to run

npx skills add https://github.com/IsKenKenYa/Commory --skill caveman-iskenkenya

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: caveman
Source: https://github.com/IsKenKenYa/Commory/tree/main/.windsurf/skills/caveman
Command: npx skills add https://github.com/IsKenKenYa/Commory --skill caveman-iskenkenya

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The caveman skill reduces token usage and response verbosity while preserving full technical correctness, making it easier to communicate when brevity matters.

Core Features & Use Cases

Ultra-compressed replies: Drops filler, articles, pleasantries, hedging, and unnecessary conjunctions while keeping technical terms exact.
Deterministic formatting style: Uses terse phrasing and a consistent pattern like “[thing] [action] [reason]. [next step].”
Safety-focused auto-clarity: Temporarily preserves clarity for security warnings, irreversible confirmations, multi-step sequences, or when the user asks to clarify.

Quick Start

Reply with: “Use caveman mode and explain how auth token refresh works in a few lines.”

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 responses without losing technical accuracy?▼

To reduce token usage without losing accuracy, apply a terse formatting style that strips filler, pleasantries, and hedging while preserving exact technical terms and code blocks. This approach compresses communication by dropping unnecessary conjunctions and articles.

What is caveman mode for technical writing and debugging?▼

Caveman mode is a deterministic formatting style for technical writing and debugging that outputs terse smart-caveman phrasing. It follows a consistent pattern like “[thing] [action] [reason]. [next step]” to compress troubleshooting steps without losing accuracy.

Can I use terse prompt formatting for stepwise troubleshooting instructions?▼

Yes, you can use terse prompt formatting for stepwise troubleshooting. It compresses instructions by removing unnecessary language, but it temporarily expands clarity for multi-step sequences, security warnings, or irreversible actions to ensure safe execution.

Does brief AI communication affect code blocks and technical terms?▼

Brief AI communication does not affect code blocks or technical terms. The compression process strictly removes filler language and hedging, while code blocks remain unchanged and exact technical terms are fully preserved for correctness.

When should I not use ultra-compressed replies for AI communication?▼

You should not use ultra-compressed replies when handling security warnings, irreversible confirmations, or multi-step sequences. The system automatically preserves clarity in these situations, temporarily expanding responses to prevent critical errors.