phd-deepread

Structured Obsidian literature notes are generated by extracting text via PyMuPDF and applying AI-assisted templates. This skill outputs markdown files with YAML frontmatter, Dataview callouts, and wikilinks, automating the extract-generate workflow for academic PDFs.

Yes, scanned documents and image-based pages are supported via an OCR fallback. The skill uses PyMuPDF for fast text-first extraction, automatically falling back to Tesseract OCR when encountering complex or scanned pages that lack embedded text layers.

A JSON canvas for critical thinking is a structured nine-node visual layout that facilitates deep critique and reasoning. This skill generates these canvases to help users systematically deconstruct research papers, providing a structured framework for analyzing academic arguments and findings.

Yes, batch processing of multiple PDFs is supported for systematic literature reviews. The skill includes an end-to-end extract-generate-canvas workflow that applies to both single papers and batch processing, streamlining the extraction and note generation across multiple academic documents simultaneously.

Yes, you need Tesseract installed along with the PyMuPDF, pytesseract, and Pillow Python libraries. These dependencies are required for the core PDF extraction functionality, with pytesseract providing the OCR fallback for scanned academic documents.

Generating YAML frontmatter and Dataview callouts from a research paper requires running the generate command after PDF text extraction. The skill applies template-driven generation to produce Obsidian-compatible markdown notes, converting extracted text into structured metadata and visual callouts.