Ultra-compressed communication mode. Cuts token usage 75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or…
Print a one-time token usage snapshot for the current Claude Code session. Shows context %, cost, cache efficiency, degradation risk. Use when you want a quick token check without opening a second terminal. Triggers on: /gauge, "check tokens", "how much context", "token snapshot".
Cross-check a Claude output using a second model (Gemini) for factual accuracy. Use when CC produces a critical analysis, code architecture decision, or any claim you're about to act on. Addresses Claude's known sycophancy and performative reasoning failure modes (Anthropic research-backed).