caveman

To compress text communication and reduce token usage, you can use a simplified language mode that maintains technical accuracy while cutting token consumption by up to 75%. This allows AI responses to remain precise without unnecessary verbosity.

No, compressing text for token efficiency does not lose technical accuracy. The system applies simplified language rules strictly for verbosity reduction while explicitly preserving all technical details and contextual correctness required for software engineering tasks.

You trigger compressed communication mode by typing "/caveman" or saying "caveman mode". It can also auto-trigger when token efficiency is requested or specific phrases are used, persisting for every response until manually stopped.

The intensity levels for AI text compression range from lite to ultra. These levels dictate the aggressiveness of the language simplification, allowing you to control the exact trade-off between maximum token reduction and baseline readability.

You should not use simplified language for AI responses when maximum readability, formal documentation, or client-facing communication is required. The aggressive token reduction relies on condensed phrasing that may appear unprofessional in standard contexts.