paper-research

To search for academic papers across arXiv, OpenAlex, and Semantic Scholar simultaneously, this skill queries all three databases in one step, deduplicates the results, and returns structured metadata including titles, authors, year, abstracts, and citation counts.

Yes, you can filter academic search results by recency and citation count. The workflow supports applying these filters to the aggregated multi-source results to help you identify the most relevant and impactful papers for your literature review.

To download PDFs and extract text for vault notes, you first review a mandatory user-guided shortlist of search results. Once you approve specific papers, the skill fetches the PDFs and extracts the text to generate structured vault notes and synthesis links.

Yes, marker-pdf is a required dependency for this skill. It is utilized within the Python scripts to handle PDF text extraction and processing when generating structured vault notes from the approved academic papers.

The best way to compile a reading list from multiple academic databases is to use a workflow that aggregates, deduplicates, and presents a shortlist for mandatory user approval before fetching PDFs, ensuring you only download the papers you actually need.

When querying arXiv, OpenAlex, and Semantic Scholar, the workflow returns JSON outputs containing structured metadata fields such as title, authors, year, abstract, arxiv_id, doi, url, citation_count, and source.