Pdf

This handles the full PDF workflow: reading text from existing files, creating new ones from markdown or HTML, and merging or splitting documents. It wraps PyMuPDF for most operations, with fallbacks to pdftotext and pandoc depending on what you have installed. The skill shines when you need to extract content from a PDF for analysis or programmatically generate reports. One thing to note: it assumes you're working with text-based PDFs, so if you're dealing with scanned documents, you'll need to layer in OCR separately. The markdown-to-PDF path via pandoc is surprisingly clean for generating formatted output without touching ReportLab directly.