Compress model responses into terse caveman-style prose with intensity levels.
npx skills add https://github.com/AlbertLin821/AIYO_new --skill caveman-albertlin821
Please help me install this Agent Skill. Skill: caveman Source: https://github.com/AlbertLin821/AIYO_new/tree/main/.agents/skills/caveman Command: npx skills add https://github.com/AlbertLin821/AIYO_new --skill caveman-albertlin821
This Skill addresses the issue of excessive token consumption in long-running AI conversations by enforcing a highly compressed, terse communication style that preserves technical accuracy while stripping away conversational filler.
Activate the caveman skill by typing /caveman to immediately begin receiving ultra-compressed responses.
High-intent search queries and answers about installing and using this skill.
To reduce token usage, compress model responses into terse prose that strips conversational filler while preserving code blocks and technical symbols. This maintains full technical accuracy while cutting output length by 65-75%.
Caveman-style communication is an ultra-compressed prose format designed for token efficiency. It drops pleasantries and filler words while ensuring error strings and technical symbols remain untouched for debugging contexts.
Start compressing AI model responses by activating the compression skill with a simple command. This immediately enforces strict rules for dropping filler and pleasantries, returning ultra-compressed responses.
Yes, ultra-compressed communication preserves code blocks, error strings, and technical symbols. The compression mechanism strictly targets prose, ensuring all technical details required to fix code remain untouched.
Available compression intensity levels range from professional-tight to extreme classical Chinese registers. These six levels include lite, full, ultra, and classical options to support diverse communication needs and debugging sessions.
Avoid using terse prose compression when conversational context, pleasantries, or detailed explanations are required for user understanding. The extreme compression levels are optimized for technical debugging, not general communication.
CANTAGESTUDIODefine refactoring session boundaries with goals, scope, and success criteria.
IsKenKenYaCompress AI communication by removing filler and hedging to reduce token usage.
ZESCODEMonitor token consumption and suggest strategic compaction intervals for multi-agent sessions.
ojrojasCompress verbose AI responses into concise caveman-style prose.
