Paper Close-Reading Notes for Obsidian

Problem Solved

Reading papers often fails not because the text is missing, but because the structure is scattered: motivation, method, experiments, and conclusions remain embedded in a PDF, while terminology and references are hard to turn into durable notes. deeppapernote targets single-paper reading and converts copyable text into structured Obsidian notes, reducing manual effort between paper and personal knowledge base.

How It Works

It follows a fixed workflow:
- Receive paper text or a file path, plus optional metadata such as title, authors, year, venue, DOI, and tags.
- Extract metadata from the beginning of the document, using placeholders like [needs verification: title] when uncertain.
- Parse sections such as Introduction, Methods, Results, Discussion, and Conclusion.
- Extract the research question, core method, key experiments, main results, conclusions, and contributions.
- Build a terminology list and an extended reading list, then assemble the note with YAML frontmatter, tags, Wiki links, internal links, and Callout blocks.
It then outputs the file path and a brief structure preview, prompting users to review uncertain fields.

Boundaries and Caveats

It fits researchers, graduate students, Obsidian users, and writers organizing literature notes or dissecting complex methods. According to the provided skill description, it does not process multiple papers in one run, does not produce full translation, does not replace reading the original paper, and does not guarantee exactness of experimental data or citation details; scanned PDFs should be converted to copyable text first. Final claims about data, conclusions, and citations should be checked against the source.