# Authorship and automated assistance JD Jones is the human author and responsible maintainer of this project. OpenAI Codex agents using GPT-6 developed mathematical arguments, Lean proofs, documentation, and verification tooling under his direction. GPT-6 Pro supplied a mathematical review, proposed the maximum inverse estimate and optimal window cutoffs, and supplied an independent exact arithmetic implementation. Codex agents adapted that implementation and formalized and checked the extensions. Gemini-Argon contributed preliminary mathematical discussion. These systems are credited as automated assistance, not as human authors or maintainers. Lean tactics construct proof terms for kernel checking. No finite numerical sample, external language-model answer, or unproved mathematical assertion serves as a proof oracle. The exact Python replays are additional finite checks and are not proofs of unrestricted quantified statements. Mechanical verification, mathematical review, worldwide priority, and registry registration are separate matters. No independent human specialist peer review or worldwide-priority determination is claimed. The formal contract and exact-commit verification evidence are described in [Verification.md](Verification.md).