aris-proof-writer

Writes rigorous mathematical proofs for ML and AI theory claims.

1.1k|116|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/OpenLAIR/dr-claw --skill aris-proof-writer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aris-proof-writer
Source: https://github.com/OpenLAIR/dr-claw/tree/main/skills/aris-proof-writer
Command: npx skills add https://github.com/OpenLAIR/dr-claw --skill aris-proof-writer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Drafting proofs for theorems, lemmas, and propositions in ML/AI research often leads to hidden gaps, unstated assumptions, or overclaimed results. This Skill produces mathematically honest proof packages that either prove the claim, correct it, or explain 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 Package: Generates a 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, missing lemma, or counterexample instead of fabricating steps.
  • Use Case: You have a convergence lemma for your optimization paper with only a rough sketch. Ask the Skill to formalize it, and it will verify the assumptions, build the dependency map, and write a complete appendix-ready proof—or tell you the claim needs weakening.

Quick Start

Ask the assistant to prove the stated theorem under the given assumptions and write the result into PROOF_PACKAGE.md.

Frequently Asked Questions about aris-proof-writer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a rigorous proof for a machine learning theorem?

State the exact claim, list all assumptions separately, and choose a strategy such as induction, contradiction, or inequality chaining. This Skill normalizes the statement, builds a dependency map of lemmas, and writes numbered justified steps into a proof package file.

How to check whether a claimed proof is actually valid?

Verify that the conclusion follows from the stated assumptions, that no cited theorem is used outside its conditions, and that no counterexample or quantifier failure exists. The Skill performs this feasibility triage and assigns a status before writing any proof.

What happens if my theorem statement is too strong to prove?

The Skill does not fabricate a proof. It explains why the original statement is unjustified, proposes a corrected weaker claim with minimal extra assumptions, and proves the corrected version instead.

Can it fill in missing steps of an incomplete proof sketch?

Yes. Provide the partial proof or sketch, and the Skill extracts the intended strategy, isolates substantial steps as separate lemmas, and completes the argument with justification for every nontrivial implication.

Where does the proof output get written?

By default the proof package is written to PROOF_PACKAGE.md in the project root, or to a file path you explicitly specify. It does not modify paper or appendix .tex files unless you explicitly request that target.