Compress AI-generated text into caveman-style prose with adjustable intensity levels.
npx skills add https://github.com/AhmadGehad/lfms --skill caveman-ahmadgehad
Please help me install this Agent Skill. Skill: caveman Source: https://github.com/AhmadGehad/lfms/tree/main/.kilocode/skills/caveman Command: npx skills add https://github.com/AhmadGehad/lfms --skill caveman-ahmadgehad
The caveman Skill solves the problem of excessively verbose AI responses by compressing them into a more concise, caveman-style prose, significantly reducing the number of tokens used while maintaining full technical accuracy.
lite to wenyan-ultra for different compression needs.Enter "caveman full" to enable full compression mode for your AI interactions.
High-intent search queries and answers about installing and using this skill.
You can reduce AI token usage by compressing text into a caveman-style prose. This approach cuts token consumption by 75% while fully preserving technical details and critical information.
To compress AI responses for an entire session, enter "caveman full". This enables persistent compression mode that automatically applies text reduction until you explicitly stop it.
Text compression does not affect security warnings because the system features auto-clarity. It automatically switches to normal prose for critical information to ensure security and clarity.
You can adjust text compression intensity using six levels ranging from lite to wenyan-ultra. These levels allow you to control the exact amount of token reduction applied to AI responses.
You should not use caveman-style text compression when maximum readability is required. However, the auto-clarity feature automatically bypasses compression for security warnings and critical technical details.
Bonny94ITACompress AI-generated text into caveman-style prose with adjustable intensity levels.
OpalBoltCompress responses by removing articles and fillers to reduce token usage.
makerjackieAutomate persistent browser interactions for testing, form filling, screenshots, and data extraction.
jwiegleyCompress natural language prompts by removing stop words while preserving semantic content.