ocr-and-documents

Extract text from PDFs and scanned documents via pymupdf or marker-pdf OCR.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/HugoAlmeidaMoreira/zeus-agent --skill ocr-and-documents-hugoalmeidamoreira
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ocr-and-documents
Source: https://github.com/HugoAlmeidaMoreira/zeus-agent/tree/main/skills/productivity/ocr-and-documents
Command: npx skills add https://github.com/HugoAlmeidaMoreira/zeus-agent --skill ocr-and-documents-hugoalmeidamoreira

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Extract text from PDFs, scanned documents, and image files using lightweight extraction (pymupdf) or high-accuracy OCR (marker-pdf), enabling quick digitization and searchability.

Core Features & Use Cases

  • OCR for scanned PDFs and images: convert non-selectable text into searchable, editable text.
  • Local and URL-based extraction: choose fast local parsing or robust OCR for difficult layouts, including batch processing.
  • Output formats: Markdown, plain text, or structured metadata extraction for integration into pipelines (e.g., research notes, archiving).

Quick Start

Provide a document (PDF, scanned image, or EPUB) and specify the desired output format to start extraction.

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 using OCR?▼

PyMuPDF handles lightweight, fast text extraction from native PDFs with selectable text. For scanned documents and images requiring OCR, the pipeline applies marker-pdf to achieve high-accuracy recognition and convert non-selectable text into editable content.

Can I extract tables and metadata from PDFs along with the main text?▼

Yes, you can extract tables and metadata alongside main text. The script-based outputs generate markdown, plain text, and structured metadata, enabling direct integration into research notes, contract digitization, or archival data retrieval pipelines.

How do I convert a PDF to markdown for academic research or archiving?▼

You convert a PDF to markdown by providing the document and specifying markdown as the output format. The extraction pipeline processes native PDFs and OCR-processed scans, yielding structured markdown suitable for academic research and archiving.

What is the best way to digitize scanned documents for searchable content?▼

The best way to digitize scanned documents for searchable content is using high-accuracy OCR via marker-pdf. This process converts non-selectable text from image-based scans into editable, searchable formats like markdown and plain text.

Does this PDF extraction pipeline support batch processing and URL-based inputs?▼

Yes, the PDF extraction pipeline supports batch processing and URL-based inputs. You can choose fast local parsing for native PDFs or robust OCR for difficult layouts, extracting text directly from local files or URLs.

What are the limitations of using lightweight extraction versus OCR for PDF text extraction?▼

Lightweight extraction via pymupdf is fast but limited to native PDFs with selectable text. OCR via marker-pdf handles scanned documents and difficult layouts but requires more processing, making it slower for simple digitization tasks.