ocr-and-documents

To extract text and tables from a scanned PDF, this Skill uses marker-pdf for OCR extraction, handling advanced layouts like tables, equations, and forms while producing structured Markdown output.

Converting an arXiv PDF into Markdown is done via the URL-first extraction feature, which fetches the document directly through web extraction and outputs clean text without requiring local file dependencies.

Yes, you can extract text from text-based PDFs using the lightweight pymupdf extraction path, which reliably parses text, tables, and metadata into Markdown format without invoking heavy OCR dependencies.

Yes, this document text extraction approach supports EPUB and HTML-like inputs alongside text-based PDFs, scanned OCR PDFs, and arXiv URLs, outputting structured Markdown or JSON with metadata.

PDF parsing fails on scanned documents using basic extractors because they lack OCR capabilities; this Skill solves the problem by routing scanned files to marker-pdf to process advanced layouts and reading order.

You need OCR for document parsing when working with scanned PDFs or image-heavy files, whereas standard text extraction works for text-based PDFs; this Skill automatically selects the appropriate method based on document type.