Compresses AI responses by removing filler words while preserving technical accuracy.
npx skills add https://github.com/GregM1991/skills --skill caveman-gregm1991
Please help me install this Agent Skill. Skill: caveman Source: https://github.com/GregM1991/skills/tree/main/skills/caveman Command: npx skills add https://github.com/GregM1991/skills --skill caveman-gregm1991
< not <=." instead of paragraph-length explanations. ## Quick Start Say "caveman mode" or "be brief" to switch the assistant into ultra-compressed responses for the rest of the conversation.High-intent search queries and answers about installing and using this skill.
Activate caveman mode by saying "caveman mode", "be brief", or "less tokens". The assistant then drops articles, filler, and pleasantries, cutting token usage by roughly 75% while keeping technical content exact.
Say "stop caveman" or "normal mode" to deactivate it. The mode persists across every response until explicitly turned off, so it will not revert on its own after multiple turns.
No. Code blocks remain unchanged, technical terms stay exact, and errors are quoted verbatim. Only prose around the technical content is compressed into fragments and abbreviations.
It temporarily drops compression for security warnings, irreversible action confirmations, multi-step sequences where fragment order risks misreading, or when you ask for clarification. Caveman style resumes after the clear part is done.
Fragmented phrasing can be ambiguous for nuanced explanations, multi-step instructions, or users unfamiliar with abbreviations like DB, auth, or req. For those cases, the auto-clarity exception or normal mode is more appropriate.
giovi1998Compress assistant responses into terse caveman-style prose with exact technical terms.
X-School-AcademyApply automated code fixes to Dart projects via dart fix --apply.
marinvchCompress long agent conversations into anchored structured summaries preserving session context.
EntelligentsiaConstrain and validate LLM outputs to a defined JSON schema.