Falsify 2 · 263311487-ux The scientific thinking protocol for AI agents. Use when facing complex, ambiguous, or high-stakes questions where guessing is costly — technical design decisions, architecture choices, debugging theories, data claims, security judgments, or any answer the agent is tempted to give confidently without evidence. Falsification first: hypothesis → attempt to break it → evidence → calibrated conclusion. Also triggers on "think like a scientist", "先别急着下结论", "证伪一下", "doubt check", "what if I'm wrong", "challenge my thinking", "what am I missing", "stress test this".
