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/biostudies.md

BioStudies

EMBL-EBI archive for the data outputs of a life-science study: files hosted here, plus links out to ArrayExpress, BioImages, ENA, and other archives. Use it when the user wants the dataset behind a paper, a BioStudies accession (S-BSST…, S-EPMC…, S-CMO…, E-MTAB…), or "supplementary data at EBI." It is not a paper index.

Find the paper in PubMed / Europe PMC, then come here with an accession or a keyword that appears in the study record.

The dated page-mode examples below are historical observations. Cursor pagination was verified against current help and a public two-page query on 2026-09-30.

Base URL

https://www.ebi.ac.uk/biostudies/api/v1

Authentication

None.

Rate Limits

No published per-second cap. Serialize. EBI asks for reasonable use.

Key Endpoints

1. Search studies

GET /search?query={text}&page={n}&pageSize={n}
GET /search?query=organoid&pageSize=2

Verified: HTTP 200 with

| Field | Value on this call | Meaning | |---|---|---| | page | 1 | 1-based | | pageSize | 2 | | | totalHits | 4195, then 4480 on a later call | Approximate | | isTotalHitsExact | false | Do not reconcile as if this were Europe PMC hitCount | | nextCursor | null in page mode | Start with pagination=cursor for a cursor walk | | hits | 2 study summaries | |

A hit has accession, type (study), title, author, files (count), release_date, isPublic, content (a flattened text blob). author may be an empty string.

totalHits moved by hundreds between two calls a few seconds apart, and isTotalHitsExact stayed false. Report "about N studies" and the page you fetched. Do not claim a complete walk against that number.

Use page=2 for normal browsing. For exhaustive retrieval, use the documented cursor mode:

curl -fsS --get 'https://www.ebi.ac.uk/biostudies/api/v1/search' \
  --data-urlencode 'query=organoid' \
  --data-urlencode 'pagination=cursor' \
  --data-urlencode 'pageSize=2'

Send the opaque returned nextCursor as cursor on the next request, keeping the same query and page size. Stop at nextCursor: null; do not construct or decode cursors, or stop based on an approximate totalHits. The official example uses pageSize=1000 for bulk walks; use a small bound for interactive work. This traverses the live index, not a frozen snapshot, so record access times and deduplicate accessions.

2. One study

GET /studies/{accession}
GET /studies/S-CMO2844

Verified: HTTP 200. The body is not the search-hit shape:

{
  "accno": "S-CMO2844",
  "type": "submission",
  "attributes": [{"name": "Title", …}, {"name": "ReleaseDate", …}],
  "section": { "type": "Study", "accno": "s1", "attributes": […], "subsections": […] }
}

Title lives in attributes (name Title), not title. Files and links are nested under section.subsections. Walk that tree; do not expect files: [ …urls ] at the top level.

A 404 means no such public accession.

Typical Workflow

  1. Search with a paper title, accession, or biological keyword.
  2. Take accession from hits[].
  3. GET /studies/{accession} for the submission tree and file list.
  4. Cite the accession and the BioStudies URL (https://www.ebi.ac.uk/biostudies/studies/{accession}).
  5. If they wanted the paper, go back to PubMed / Europe PMC with the title or a DOI found in the study attributes.

Failure Modes

| What you did | What happens | What to do | |---|---|---| | Treated totalHits as exact | Count drifts; isTotalHitsExact is false | Say "about N"; do not exit-4 reconcile | | Expected search fields on /studies/{acc} | No title, no files count | Read attributes and section | | Used BioStudies as PubMed | Studies, not articles | Search literature APIs first | | Expected a cursor in page mode | nextCursor is null | Start with pagination=cursor; then send cursor |

Official sources reviewed 2026-09-30

  • https://www.ebi.ac.uk/biostudies/help#api