ocr-and-documents

To extract text from a scanned PDF using OCR, this Skill applies marker-pdf to recognize text, equations, and complex layouts, producing clean searchable Markdown output. It routes scanned documents to this advanced parsing path for high accuracy across many languages.

The best way to convert an arXiv PDF to Markdown is using this Skill's web extraction path for URLs or lightweight local parsing for text-based files. This workflow extracts searchable text and structured tables from research papers for downstream summarization and review.

Yes, marker-pdf works better than pymupdf for complex document parsing when you need OCR, equation recognition, form extraction, or better reading-order detection. The Skill selects pymupdf for standard text-based PDFs and marker-pdf for advanced layout requirements.

Yes, you can extract tables from PDF files for research notes using this Skill's Markdown and table extraction features. It processes native PDFs and scanned documents to isolate structured tables and relevant text sections for your downstream research workflow.

You should use web extraction for PDF documents when you have a URL available, and local parsing when processing local files. The Skill requires selecting web_extract for URLs, otherwise it uses local pymupdf or marker-pdf extraction based on OCR and layout needs.

No, you do not need a separate OCR tool for scanned documents with complex layouts because this Skill uses marker-pdf internally. It handles OCR, equations, forms, and complex layout tasks directly, providing strong accuracy without requiring external OCR software.