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/opencitations.md

OpenCitations

Open citation data (who cites whom) as open lists of citing/cited PIDs. Use it when you need an openly licensed citation edge, a count you can cite, or an Open Citation Identifier (OCI). It is not a paper search and it does not return titles on the Index endpoints.

For a literature search use PubMed, OpenAlex, or Semantic Scholar. For a citation graph with titles and abstracts, start with Semantic Scholar or OpenAlex and treat OpenCitations as the open-data check. Europe PMC /citations is the biomedical alternative when you already have a {source}/{id} pair.

All figures below verified 2026-09-10.

Base URLs

https://api.opencitations.net/index/v2    # citation edges and counts
https://api.opencitations.net/meta/v1     # bibliographic metadata for a PID

Index v2 is current (v2.2.0, 2025-04-15). Meta lives at v1 — meta/v2/... is HTTP 404.

Authentication

Optional. Public calls work without a token. For heavier use, request an access token from OpenCitations and send Authorization: <token>. Do not add a new .env key for this; proceed without one.

Rate Limits

The Index API documents 180 requests/minute per IP. Serialize requests. Call /citation-count before /citations — the list endpoint returns every incoming citation in one body, with no page parameter.

Identifier prefix (required)

Index v2 accepts doi:, pmid:, omid:, issn:, and isbn:; for papers prefer the first three. A bare DOI is HTTP 400:

GET /index/v2/citation-count/10.1038/nature12373
  -> 400  the value '10.1038/nature12373' is not valid for parameter 'id'
          Example: /index/v2/citation-count/doi:10.1108/jd-12-2013-0166

GET /index/v2/citation-count/doi:10.1038/nature12373
  -> 200  [{"count": "1806"}]

Key Endpoints

1. Incoming citation count

GET /index/v2/citation-count/{id}

Always a one-element JSON array. count is a string, not an integer.

| Query | HTTP | Body | |---|---|---| | doi:10.1038/nature12373 | 200 | [{"count": "1806"}] | | pmid:23803767 | 200 | [{"count": "94"}] | | doi:10.9999/not-a-real-doi | 200 | [{"count": "0"}] |

A missing work is HTTP 200 with "0", not 404. Do not treat 200 as "this DOI is in the index."

2. Outgoing reference count

GET /index/v2/reference-count/{id}

Same shape as citation-count.

3. Incoming citations / outgoing references

GET /index/v2/citations/{id}
GET /index/v2/references/{id}

Each item:

| Field | Meaning | |---|---| | oci | Open Citation Identifier (citingOmidsuffix-citedOmidsuffix) | | citing / cited | Space-separated PIDs, each prefixed (doi:, pmid:, omid:, openalex:) | | creation | ISO date of the citing work | | timespan | XSD duration (P6Y0M1D) between cited and citing publication | | journal_sc / author_sc | "yes" / "no" self-citation flags |

Verified on doi:10.1038/nature12373 /references: 30 rows. First citing is omid:br/06120344846 doi:10.1038/nature12373 openalex:W2159974629 pmid:23903748. Parse the doi: token out; do not take the whole string as one DOI.

Verified on doi:10.1186/1756-8722-6-59 /citations: 217 rows in one response. For nature12373 the count is 1806 — do not pull that list unless the user asked for the full set.

4. One citation by OCI

GET /index/v2/citation/{oci}

oci is the two-number form without an oci: prefix.

5. Metadata (titles, authors)

GET /meta/v1/metadata/{id}

Same doi: / pmid: / omid: prefix. Returns id (space-separated PIDs), title, author (semicolon-separated, may include ORCID + OMID). Use this when you have an edge from Index and need a human-readable label.

Typical Workflow

  1. You have a DOI or PMID.
  2. citation-count first. If "0", say OpenCitations has no incoming citations — not that the paper is uncited everywhere.
  3. For a short list, /references or /citations. Extract the doi: token from each citing/cited string.
  4. Resolve titles with Meta, Crossref, or Semantic Scholar. Do not invent them from the OCI.

Failure Modes

| What you did | What happens | What to do | |---|---|---| | Bare DOI, no doi: prefix | HTTP 400 | Prefix the scheme | | Unknown DOI | HTTP 200, count: "0" | Report a gap; try Semantic Scholar | | Parsed citing as one DOI | You store omid:br/… doi:10.… pmid:… | Split on spaces; keep the doi: value | | /citations on a highly cited work | Multi-megabyte JSON, no pagination | Count first; bound the pull | | meta/v2/... | HTTP 404 HTML | Use meta/v1 |

Official sources reviewed 2026-09-30

  • https://api.opencitations.net/index/v2
  • https://api.opencitations.net/meta/v1