What problem does it solve? Long, filler-heavy AI responses waste tokens and slow down reading. This Skill cuts output tokens by roughly 65% (measured) by stripping articles, filler, pleasantries, and hedging while keeping every technical detail, code snippet, and error string exact. ## Core Features & Use Cases - Six intensity levels: lite, full (default), ultra, plus wenyan-lite, wenyan-full, and wenyan-ultra for classical Chinese compression, switchable mid-session with /caveman lite|full|ultra. - Language preservation: compresses the user's dominant language (Portuguese, Spanish, Chinese, etc.) instead of forcing English, and never alters code, API names, CLI commands, or error strings. - Auto-clarity override: automatically drops compression for security warnings, irreversible-action confirmations, and multi-step sequences where fragments could cause misreads. - Use Case: During a long debugging session, invoke /caveman so every diagnosis arrives as terse fragments like "Bug in auth middleware. Token expiry check use < not <=. Fix:" — then say "normal mode" to revert. ## Quick Start Ask the AI to enable caveman mode for the rest of the session so all answers come back ultra-terse with full technical accuracy.
