ocr-and-documents

Extract text, tables, and metadata from PDFs and scanned documents.

3|Updated Apr 4, 2026
One-click install
npx skills add https://github.com/twjarviszyra-web/hermes-unbound --skill ocr-and-documents-twjarviszyra-web
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ocr-and-documents
Source: https://github.com/twjarviszyra-web/hermes-unbound/tree/main/skills/productivity/ocr-and-documents
Command: npx skills add https://github.com/twjarviszyra-web/hermes-unbound --skill ocr-and-documents-twjarviszyra-web

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires marker-pdf, pymupdf, pymupdf4llm, and includes scripts (resource) components.

What problem does it solve?

It removes the manual burden of turning PDFs, scans, images, and office documents into usable text, so you can quickly read, search, quote, and analyze content that would otherwise be trapped in files.

Core Features & Use Cases

  • Remote URL extraction: Prefer web-based extraction when a document is available online.
  • Local PDF parsing: Use lightweight parsing for standard text PDFs and OCR for scanned or image-heavy files.
  • Document coverage: Handle PDFs, DOCX, images, EPUB, tables, page ranges, and metadata.
  • Practical use cases: Research papers, invoice capture, scanned archive digitization, and extracting tables or figures from reports.

Quick Start

Ask the assistant to extract the text, tables, and metadata from your attached document, using the URL first when available and OCR for scanned pages.

Frequently Asked Questions about ocr-and-documents

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

FAQPage Schema
How do I extract text from a scanned PDF for research papers?

To extract text and tables from PDFs, attach your document or provide a URL. The tool prioritizes web extraction for online sources and uses local parsing or OCR for local files to output markdown, metadata, and tables.

Does this PDF extraction method work for capturing tables from reports?

Yes, PDF extraction supports table capture from reports and invoices. It uses local parsing and marker-pdf tooling to identify and extract tables alongside text, images, and metadata into editable markdown formats.

What is the best way to digitize scanned documents into searchable text?

The best way to digitize scanned documents is using OCR to convert them into searchable, editable text. This applies to scanned archives, images, and image-heavy PDFs, yielding markdown output with captured metadata.

Can I extract metadata and specific page ranges from a local PDF?

Yes, you can extract metadata and specific page ranges from local PDFs. The tooling uses lightweight parsing and marker-pdf to isolate page ranges, capture images, and output structured markdown with metadata.

When should I use web extraction versus local parsing for document conversion?

Use web extraction when a document is available online via a remote URL, and use local parsing or OCR for local files, scanned documents, and image-heavy PDFs that need document conversion into markdown.

What are the limitations of using lightweight PDF parsing for scanned archives?

Lightweight PDF parsing cannot read text embedded in images, so it fails on scanned archives. You must use OCR tooling to accurately extract text from image-heavy or scanned document files.