book-analyzer

To extract text from EPUB and PDF books, this tool uses pandoc for EPUB files and pdftotext for PDFs, splitting the content into chapters and generating structured notes with key concepts. If extraction quality is poor, it falls back to reading order.

The best way to generate chapter-by-chapter summaries from an EPUB file is using parallel agent workflows that analyze each chapter individually, extracting quotes, examples, and questions before synthesizing the book's core thesis.

Yes, you need pandoc and pdftotext installed to analyze PDF and EPUB books, as these dependencies handle the text extraction. Python3 and ebook-convert are also required to run the extraction pipeline.

Yes, you can automatically create Obsidian notes with wikilinks from a local PDF. The tool synthesizes cross-chapter arguments and writes vault-ready markdown files, including optional term notes for key concepts.

If pdftotext fails to extract clean text from your PDF book, the tool validates the extraction quality and falls back to reading the document directly, ensuring you still get structured chapter notes.

Cross-chapter synthesis works by orchestrating parallel agents that analyze individual chapters, then combining those results to map the book's argument architecture and extract the core thesis into vault-ready markdown.