skills.rest
danielsuguimoto
danielsuguimotoCOMMUNITY
Explore all skills by this author →
𝕏@djsuguimotoGitHub

caveman

Compresses AI responses by removing filler, hedging, and narration while preserving technical accuracy.

Updated Jul 11, 2026
Search Tags:#communication style#token efficiency#brevity#response formatting#text compression#concise writing
One-click install
npx skills add https://github.com/danielsuguimoto/skills --skill caveman-danielsuguimoto
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: caveman
Source: https://github.com/danielsuguimoto/skills/tree/main/skills/caveman
Command: npx skills add https://github.com/danielsuguimoto/skills --skill caveman-danielsuguimoto

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI responses often waste tokens and reading time on filler phrases, hedging, pleasantries, and tool-call narration. This Skill enforces a compressed communication style that keeps full sentences and technical precision while cutting everything else. ## Core Features & Use Cases - Filler and hedging removal: Strips filler words, preambles, recaps, apologies, and success confirmations from every reply. - Accuracy preservation: Keeps code blocks, API names, CLI commands, error strings, and commit prefixes verbatim. - Stealth execution: Suppresses tool-call narration and progress recaps, speaking only for results, blockers, or questions. - Use Case: During a long debugging session, activate caveman mode so every diagnosis and fix instruction arrives as short, direct prose instead of verbose explanations, reducing token usage across the session. ## Quick Start Tell the agent to use caveman mode for the rest of this session to get brief, token-efficient replies.

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 more concise?▼

Activate caveman mode by saying "caveman mode", "be brief", or "less tokens". The skill removes filler, hedging, and narration from every reply while keeping full sentences and technical terms intact.

How to reduce token usage in AI chat responses?▼

Use caveman mode to compress both visible replies and reasoning tokens. It cuts preambles, recaps, and acknowledgments, which typically account for a large share of response length in long sessions.

Does concise mode affect code blocks and error messages?▼

No. Code blocks, API names, CLI commands, commit prefixes, and exact error strings are preserved verbatim. Compression applies only to conversational prose, never to technical content.

When does caveman mode switch back to normal writing?▼

It automatically uses normal prose for security warnings, irreversible action confirmations, and multi-step sequences where compression risks misreading. Saying "stop caveman" or "normal mode" turns it off entirely.

Does compressed writing apply to commits and pull requests?▼

No. Code, commits, PRs, and content posted to external tools are written normally. Compression applies only to internal conversation replies and reasoning, not to anything published externally.

Related Skills

Explore More Content & Communication Skills→
leafcoderleafcoder
2

user-customization

Manage a centralized configuration file for AI agent technical preferences and communication styles.

Community
Basic
gonzalezpazmonicagonzalezpazmonica
47

headroom-optimization

Compress AI context windows using a 5-phase framework to reduce token usage.

Community
Intermediate
adiytharpansaadiytharpansa

response-optimizer

Review drafted messages for accuracy, clarity, completeness, and effectiveness.

Community
Intermediate
tabesinktabesink

bmad-distillator

Distill source documents into lossless, LLM-optimized bullet-only distillates.

Community
Advanced