proof-finder

Extract source-indexed proof material with stable IDs from papers and notes.

2|1|Updated May 7, 2026
One-click install
npx skills add https://github.com/shenmuxing/EtaSkill --skill proof-finder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: proof-finder
Source: https://github.com/shenmuxing/EtaSkill/tree/main/skill-examples/proof-finder
Command: npx skills add https://github.com/shenmuxing/EtaSkill --skill proof-finder

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you turn proof-heavy papers, notes, PDFs, Markdown, or LaTeX sources into source-indexed proof material instead of loose notes, making technical lemmas and strategies easier to reuse later.

Core Features & Use Cases

  • Proof Material Extraction: Identifies theorems, lemmas, definitions, dependencies, reductions, and proof strategies from a source document.
  • Source-Indexed Output: Preserves paper locations, stable material IDs, and reusable abstractions for later organization.
  • Verification Workflow: Uses DeepSeek for first-pass screening and blind backtests, then has Codex verify against the source before writing entries.
  • Use Case: If you are reading a dense paper and want to capture only the strongest reusable proof ingredients, this Skill helps you mine them into structured material files without prematurely classifying them into final taxonomy categories.

Quick Start

Use the proof-finder skill to extract source-indexed proof material from the provided paper and update the material index.

Frequently Asked Questions about proof-finder

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

FAQPage Schema
How do I extract reusable proof strategies and lemmas from a PDF research paper?

You can extract proof material from PDFs by identifying theorems, lemmas, and dependencies, then saving them into source-indexed files with stable IDs and paper locations for later reuse.

What is proof mining and how does it preserve lemma dependencies from LaTeX or Markdown sources?

Proof mining extracts reusable proof strategies and technical lemmas from LaTeX or Markdown. It preserves lemma dependencies and source locations, outputting structured material files instead of loose notes.

Does extracting proof material from dense papers require DeepSeek and Codex verification?

Yes, extracting proof material requires DeepSeek for first-pass screening and blind backtests, followed by Codex verification against the source document before updating any proof-material outputs.

Can I capture theorem proving logic from LaTeX without prematurely sorting it into a final taxonomy?

Yes, you can capture theorem proving logic from LaTeX by mining the strongest reusable proof ingredients into structured material files, intentionally avoiding premature classification into final taxonomy categories.

What is the best way to index source locations for mathematical proofs extracted from notes?

The best way to index source locations for extracted proofs is to generate source-indexed files that record stable material IDs and exact paper locations alongside the captured theorems and proof strategies.

Why do I need blind backtests when extracting theorems and reductions from research notes?

Blind backtests are required during theorem extraction to validate the DeepSeek first-pass screening, ensuring the mined reductions and proof strategies are accurate before Codex verifies and writes entries.