paper-synthesizer

To extract structured knowledge from a research paper PDF, attach the paper URL so the skill can ingest the document and pull out the title, authors, year, venue, and core contributions into a standardized synthesis.

Yes, you can generate a summary with metadata from an arXiv paper by providing its URL. The skill uses the arXiv ID as an identifier to produce a traceable markdown staging file for knowledge base ingestion.

The best way to synthesize academic papers for comparison is to process each PDF individually to capture its metadata and key results in a standardized staging file, enabling quick and source-traceable comparisons across papers.

No external dependencies are required to process documents. You simply provide the paper URL to fetch and read the PDF, and the skill outputs a staging file path under the local knowledge base directory.

Extracted research paper summaries are stored as markdown files with YAML frontmatter at the staging file path under the $HOME/.ata/knowledge-base/staging directory, ready for downstream ingestion.

When a DOI is not available, the skill handles paper identification by using arXiv IDs or generating title-based slugs to ensure the synthesis remains traceable within the knowledge base.