proof-writer

Generate structured mathematical proof packages for ML/AI theory claims.

2|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/chenghaoYang/auto-coder-trainer --skill proof-writer-chenghaoyang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: proof-writer
Source: https://github.com/chenghaoYang/auto-coder-trainer/tree/main/aris/skills/proof-writer
Command: npx skills add https://github.com/chenghaoYang/auto-coder-trainer --skill proof-writer-chenghaoyang

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the drafting of mathematically rigorous proofs for ML/AI theory, filling gaps, formalizing sketches, and validating argument structure.

Core Features & Use Cases

  • Structured Claim Normalization: capture exact statements and assumptions from user input.
  • Proof Strategy & Dependency Mapping: plan steps, dependencies, and lemmas needed.
  • Proof Package Generation: output a complete PROOF_PACKAGE.md structure with status, assumptions, and steps.
  • Blockage & Revision: generate blockage reports when a claim is not provable as stated.
  • Use cases: Prove theorems, lemmas, and corollaries in ML theory, or formalize a user-provided sketch into a complete proof artifact.

Quick Start

Provide the exact theorem statement and assumptions to begin the proof drafting process.

Frequently Asked Questions about proof-writer

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

FAQPage Schema
How do I write rigorous mathematical proofs for machine learning theorems?

Drafting ML theory proofs step by step requires enforcing explicit assumptions and notation, then mapping dependencies to produce a complete proof artifact with claims, statuses, and logical steps.

Can I formalize a partial proof sketch into a complete ML theory proof?

Formalizing a partial ML proof sketch involves incorporating user sketches to address gaps, enforcing explicit assumptions, and generating a complete structured proof artifact with stepwise justifications.

What is the best way to structure assumptions and notation when proving ML theorems?

Structuring ML theorem assumptions relies on structured claim normalization to capture exact statements, ensuring the resulting proof package maintains rigorous stepwise justification and clear dependency tracking.

What happens when an ML theorem is not provable as stated during formalization?

If an ML theorem is unprovable as stated, the formalization process outputs a blockage report detailing the logical barrier, requiring you to revise explicit assumptions or adjust the theorem statement.

Does formalizing ML theory proofs require prior knowledge of formal verification tools?

No formal verification tool expertise is needed to formalize ML theory proofs; the process accepts explicit assumptions and user sketches directly to produce a structured proof package with stepwise justifications.