paper-lookup

Trigger phrases

Phrases that activate this skill when typed to Claude Code:

• find papers on
• look up this DOI
• search PubMed for
• find this paper
• get the abstract for

What it does

paper-lookup is a Claude Code skill from K-Dense AI’s scientific-agent-skills repo. It turns Claude into a multi-database paper retrieval agent that queries up to 10 academic databases via REST APIs in a single session: PubMed, PMC (full text), bioRxiv, medRxiv, arXiv, OpenAlex, Crossref, Semantic Scholar, CORE, and Unpaywall.

A session produces structured results — abstracts, full-text links, open-access availability, citation graphs, and author metadata — depending on which databases have coverage. For DOI or PMID lookups the result is immediate; for topic searches the skill returns a ranked list you can narrow.

When to use it

Reach for it when:

• You have a DOI, PMID, or arXiv ID and want the full metadata record
• You’re looking for preprints on a specific topic across bioRxiv, medRxiv, and arXiv simultaneously
• You want to check whether a paywalled article has an open-access version via Unpaywall

When not to reach for it:

• You need a synthesized literature review — use literature-review
• You want to interact with full-text experimental data extracted from papers — use bgpt-paper-search

Install

Copy the SKILL.md from K-Dense AI’s paper-lookup folder into .claude/skills/paper-lookup/ in your project.

Trigger phrases: “find papers on”, “look up this DOI”, “search PubMed for”, “find this paper”.

What a session looks like

A typical session has three phases:

1. Query specification. Provide a DOI, PMID, arXiv ID, or search terms. Optionally specify which databases to query and any date filters.
2. Parallel API retrieval. Claude queries the relevant databases concurrently and de-duplicates results by DOI where records overlap.
3. Structured output. Results are returned with title, authors, abstract, journal/venue, publication date, DOI, and open-access link when available. Citation counts from Semantic Scholar or OpenAlex are included where present.

Receipts

Where it works well:

• DOI and PMID direct lookups — resolution is fast and metadata is complete for indexed records
• Cross-server preprint searches where the same work may appear on bioRxiv, medRxiv, and arXiv

Where it backfires:

• Books, book chapters, and conference proceedings have lower coverage across these APIs than journal articles
• Unpaywall links are only as current as their crawler; some recent open-access versions lag by days to weeks

Pattern that works: use paper-lookup for targeted retrieval, then hand the results to citation-management to format them for your bibliography.

Source and attribution

Originally authored by K-Dense Inc.. The canonical SKILL.md lives in the paper-lookup folder of their public scientific-agent-skills repository.

License: MIT. Install, adapt, and redistribute with attribution preserved.

This page documents the skill from a practitioner’s perspective. For the formal spec and any updates, defer to the source repo.