Compress AI responses to reduce token consumption while preserving technical accuracy.
npx skills add https://github.com/Avihusitton/gil-therapy --skill caveman-avihusitton
Please help me install this Agent Skill. Skill: caveman Source: https://github.com/Avihusitton/gil-therapy/tree/main/.agents/skills/caveman Command: npx skills add https://github.com/Avihusitton/gil-therapy --skill caveman-avihusitton
This Skill eliminates wasted token usage in AI responses caused by unnecessary filler language, pleasantries, and redundant phrasing, which drains context window space, increases processing costs, and slows down technical workflows.
Request a concise explanation of your current technical issue using caveman mode full intensity to get a direct, fluff-free response.
High-intent search queries and answers about installing and using this skill.
You can reduce AI token usage by applying ultra-compressed, fluff-free communication that eliminates unnecessary filler language while retaining full technical accuracy for code debugging and troubleshooting workflows.
Concise AI responses for pull request reviews are achieved by using compressed communication modes that strip pleasantries and redundant phrasing while preserving exact formatting for code blocks, error strings, and API references.
Yes, you can adjust response compression intensity across 6 tiers—lite, full, ultra, wenyan-lite, wenyan-full, and wenyan-ultra—to match your needed balance of brevity and clarity for technical queries.
Compressed AI communication preserves exact formatting for code blocks, error strings, and API references, ensuring technical accuracy is maintained even when response token consumption is reduced by approximately 75%.
Shorter AI responses auto-trigger when you request token efficiency, brevity, or use specific trigger phrases like "caveman mode" or "be brief" during software engineering tasks.
You should avoid ultra-compressed AI responses when maximum clarity is prioritized over token efficiency, or when your context requires extensive conversational detail rather than direct, fluff-free technical explanations.
junechen7414Compress assistant responses into terse, token-efficient prose while preserving technical accuracy.
Chris-MaskeyCompress pull request review findings into one-line comments with location, problem, and fix.
chiruu12Remove filler words from responses to reduce token usage by about 75%.
docs-plusReduce token consumption in AI technical responses by removing filler language.
