AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

paper-lookup

Searches 18 scholarly APIs for papers, preprints, citations, open-access full text, repository records, and journal OA status, and returns results with reproducible provenance. Covers PubMed, PMC, Europe PMC, bioRxiv, medRxiv, arXiv, OpenAlex, Crossref, Semantic Scholar, CORE, Unpaywall, OpenCitations, PubTator3, Zenodo, Figshare, ROR, BioStudies, and DOAJ. Use when searching for papers, citations, DOI/PMID/arXiv lookups, abstracts, full text, open-access PDFs, preprints, citation graphs, author publications, biomedical entity annotations, deposited records (Zenodo, Figshare, BioStudies), institution ROR IDs, or any scholarly literature query. Triggers on mentions of any supported database or requests like "find papers on X", "look up this DOI", "who cites this paper", or "get me the PDF".

Original source
K-Dense-AI
License
MIT
Pinned revision
5f2bc0341354fd5590aef76a533ada1e1e5b4e62

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

This skill also requires local files or dependencies. Remote reading alone may not be enough.

Ask your assistant

Use Agentic Memory skills to read k-dense-ai--paper-lookup at revision 5f2bc0341354fd5590aef76a533ada1e1e5b4e62, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 19

references/figshare.md

Figshare

A general research repository (figures, datasets, posters, papers, media). Use it when the user names Figshare or a figshare.com DOI, or wants files deposited there. It is not a journal index. For OA journal PDFs use Unpaywall; for EBI-hosted study packages use BioStudies; for CERN-style software dumps prefer Zenodo.

All figures below verified 2026-09-10.

Base URL

https://api.figshare.com/v2

Docs: https://docs.figshare.com/v2/

Authentication

Public article metadata does not need a token. Private records, uploads, and account endpoints do (Authorization: token ACCESS_TOKEN). This skill only uses the public routes.

Rate Limits

The official recommendation is at most 1 request/second. Abuse can be throttled or blocked even though the docs describe no automatic rate limiter. Serialize.

Key Endpoints

1. Search — POST, not GET

POST /v2/articles/search
Content-Type: application/json

{"search_for": "CRISPR", "page": 1, "page_size": 10}

Verified: HTTP 200, a bare JSON array (no total, no hits wrapper). First hit was a CRISPR supplementary dataset (defined_type_name: dataset).

There is no count in the body and no useful Link/X-Count header on this call. Walk page until a page comes back shorter than page_size or empty. Do not invent a total. The API limits paging depth to an offset of 1,000 (page limit depends on page_size); a 422 beyond that is a bound, not completion. Narrow a query or use the documented bulk interface for larger sets.

GET is not a search. This is the trap that produces a confident wrong paper:

GET /v2/articles?search_for=CRISPR&page_size=1

Verified: HTTP 200, one article, title Social capital in the workplace…, DOI 10.1016/j.labeco.2007.07.006. The query string is ignored; you are listing articles. If the user asked for CRISPR and you used GET, you will report the wrong object with a 200.

2. One article

GET /v2/articles/{id}
GET /v2/articles/12345

Verified: HTTP 200 with id, title, doi, defined_type_name, url, authors, files, license. Use this after search, or when the user already has a Figshare id.

Files, when public, appear on the article object. A 404 is either no such id or a record you are not allowed to read (the API uses 404 for both).

Typical Workflow

  1. POST /articles/search with a JSON body.
  2. Read id, title, doi, defined_type_name from each element.
  3. GET /articles/{id} only if you need files or a fuller record.
  4. If they wanted papers about a topic, go to OpenAlex / PubMed. Figshare search is repository search.

Failure Modes

| What you did | What happens | What to do | |---|---|---| | GET /articles?search_for=… | HTTP 200, unrelated latest-ish articles | POST /articles/search | | Expected {hits: …, total: N} | A raw array | Treat [] as empty; no total to reconcile | | Reported a Figshare hit as a journal article | defined_type_name may be dataset, figure, media | Read and report the type |

Official sources reviewed 2026-09-30

  • https://docs.figshare.com/v2/