proof-writer

Draft rigorous mathematical proofs for ML/AI theory into PROOF_PACKAGE.md.

Updated Apr 26, 2026
One-click install
npx skills add https://github.com/jandan138/Auto-claude-code-research-in-sleep --skill proof-writer-jandan138
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: proof-writer
Source: https://github.com/jandan138/Auto-claude-code-research-in-sleep/tree/main/skills/proof-writer
Command: npx skills add https://github.com/jandan138/Auto-claude-code-research-in-sleep --skill proof-writer-jandan138

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps researchers transform informal mathematical arguments into rigorous, machine-checkable proofs for ML/AI theory, ensuring clarity, completeness, and reproducibility.

Core Features & Use Cases

  • Structured proof drafting: converts a user-provided claim, assumptions, and sketch into a complete PROOF_PACKAGE.md.
  • Assumption normalization and notation handling to reduce ambiguity.
  • Proof status classification (provable as stated, provable after weakening, or not justified) with a dependency map.
  • Blockage reporting when a claim cannot be justified.

Quick Start

Provide the exact theorem statement with explicit assumptions and any sketch, and ask the agent to generate a PROOF_PACKAGE.md.

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 ML theory?

To write rigorous mathematical proofs for ML theory, provide an exact theorem statement with explicit assumptions and any sketch, and the agent generates a complete PROOF_PACKAGE.md containing step-by-step justifications and a dependency map.

What is the best way to formalize an informal proof sketch into a machine-checkable format?

Formalizing an informal proof sketch into a machine-checkable format requires normalizing assumptions and clarifying notation to reduce ambiguity, then producing a structured proof package with a dependency map and explicit justifications.

Can I use this to draft lemmas and corollaries or only full theorems?

Yes, you can use this to draft lemmas and corollaries, as well as full theorems, propositions, and proof sketches for ML/AI theory. It fills gaps and clarifies notation across all these mathematical statement types.

What happens if a mathematical claim cannot be justified during proof drafting?

If a mathematical claim cannot be justified during proof drafting, the tool issues a blockage report. It also classifies the proof status as provable as stated, provable after weakening, or not justified.

Does proof-writer require any specific dependencies or frameworks to run?

No specific dependencies or frameworks are required to run proof-writer. You simply provide the exact theorem statement with explicit assumptions and any sketch to generate the proof package.