What problem does it solve? Getting high-quality figures out of research papers is tedious: screenshots from PDFs are low-resolution, and manually hunting through arXiv source files is slow. This Skill automates figure extraction with a three-tier strategy that prioritizes original images from arXiv source packages over embedded PDF images. ## Core Features & Use Cases - arXiv Source Extraction: Downloads the e-print source tarball and pulls original PNG/JPG/EPS/SVG figures from pics/, figures/, fig/, images/, or img/ directories. - PDF Fallback Extraction: Uses PyMuPDF to extract embedded image objects or render PDF figure pages to PNG at 150 DPI when source packages are unavailable. - Indexed Output: Generates an index.md cataloging every extracted image with filename, path, size, format, and source label (arxiv-source, pdf-figure, pdf-extraction). - Use Case: Given arXiv ID 2510.24701, run the extraction script to download its source package, copy all original figures into an output directory, and produce a metadata index for a literature review or presentation. ## Quick Start Extract all figures from arXiv paper 2510.24701 into an images folder with a generated index file.
