What problem does it solve? Drafting proofs for ML/AI theory often produces polished but flawed arguments with hidden gaps, unstated assumptions, or misapplied theorems. This Skill produces mathematically honest proof packages that either prove the claim, correct it, or report exactly why it cannot be justified. ## Core Features & Use Cases - Feasibility Triage: Classifies every claim as PROVABLE AS STATED, PROVABLE AFTER WEAKENING / EXTRA ASSUMPTION, or NOT CURRENTLY JUSTIFIED before writing anything. - Structured Proof Packages: Writes a standardized PROOF_PACKAGE.md containing the exact claim, assumptions, notation, proof strategy, dependency map, numbered steps, and open risks. - Honest Blockage Reports: When a proof cannot be completed, it documents the exact blocker and the minimal extra assumption or corrected weaker statement instead of fabricating steps. - Use Case: You have a lemma in your paper appendix whose sketch only supports a weaker bound. The Skill identifies the gap, states the corrected claim with the minimal extra assumption, and writes a complete verified proof of the repaired statement. ## Quick Start Ask the AI to prove the stated theorem under the given assumptions and write the result into PROOF_PACKAGE.md.
