pymupdf-pdf

What problem does it solve? Extracting text and content from PDF documents often requires heavy OCR pipelines or slow layout-aware parsers. This Skill provides fast local PDF parsing with PyMuPDF, producing Markdown or JSON output in about one second per page when speed matters more than layout fidelity. ## Core Features & Use Cases - Fast text extraction: Convert PDFs to Markdown with page markers or to simple per-page JSON using PyMuPDF (fitz). - Optional image and table extraction: Pull embedded images into an images folder and generate rough line-based table JSON with command-line flags. - Per-document output folders: Each parsed PDF gets its own directory under a configurable output root containing output.md, output.json, images, and tables.json. - Use Case: A researcher needs to quickly extract text from a batch of arXiv papers to build a metadata pack. Running the parser on each PDF produces clean Markdown files in seconds without downloading multi-gigabyte models. ## Quick Start Ask the agent to parse a PDF file with the pymupdf_parse.py script, for example by requesting Markdown output from a specific document path into the default pymupdf-output folder.