The scientific thinking protocol for AI agents — falsify before you believe. 5-stage falsification skill (axioms → hypothesis → adversarial → verify → converge) for Codex, Claude Code, Cursor, Gemini CLI & 20+ agents. Dual-model evals 26/28. Five-stage scientific thinking protocol: axioms → hypothesis → adversarial → verify → converge.
These files are 263311487-ux/falsify's own configuration. They tell Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.
AGENTS.md A 280 tok