Crossref API — scholarly papers & DOI metadata (no key)

Search the Crossref catalog — 150M+ scholarly works (journal articles, conference papers, book chapters, preprints, datasets) — behind a clean HTTP API (and MCP tool). Search by title/keywords and author, filter by year and type, sort by relevance / citations / date; look up any DOI for full metadata; and search journals. Perfect for research assistants, citation tools, literature reviews and agents. No API key. Pay per query.

• 🔎 Search works — by query and author, filtered by year range and type, sorted by relevance, most-cited or most-recent.
• 📄 Get by DOI — full metadata for a single DOI: title, authors, journal, year, citation count, abstract (when available).
• 📚 Search journals — by name: publisher, ISSN, total DOIs.
• 🌍 Official & keyless — straight from Crossref, the DOI registration agency behind most academic publishing.
• 💵 Pay per query · no key.

The use case it was built for

Give your agent a real literature search: "find the most-cited papers on diffusion models since 2020", resolve a DOI to a clean citation, or check which journal an ISSN belongs to. Crossref indexes works from virtually every major publisher, returned as clean typed rows.

Sample output (search, query "attention is all you need", sort "cited")

{
  "ok": true, "query": "attention is all you need", "totalResults": 1383718, "count": 1,
  "results": [
    { "title": "Attention Is All You Need", "authors": ["Ashish Vaswani", "Noam Shazeer"],
      "year": 2017, "type": "proceedings-article", "container": "…",
      "doi": "10.5555/3295222.3295349", "url": "https://doi.org/10.5555/3295222.3295349",
      "citedByCount": 100000, "subjects": [] }
  ]
}

How to use

• HTTP API (Standby): POST or GET /search, /work, /journals. GET / returns help. Example: GET /search?query=diffusion%20models&sort=cited&fromYear=2020.
• Normal run: pass { operation, query | doi, ... }; results land in the dataset + key-value store.
• MCP tool: expose it to your agent via Apify's MCP server.

Input

• operation — search · work · journals.
• query — title/keywords (search) or journal name (journals). author — surname filter. doi — for work.
• type — journal-article / proceedings-article / book-chapter / posted-content / … sort — relevance / cited / published.
• fromYear / toYear — publication-year range. limit — cap rows.

Pricing

Per query from $0.005 (down to $0.0015 on higher tiers). One call = one query (returns many works or journals at once).

Related Actors

• Open Library Books — books, editions and authors.
• Wikipedia API — article search, summaries and content.

FAQ

Do I need a key? No — Crossref is keyless. This actor uses the polite pool (with a contact mailto) for reliable throughput.

Does it cover preprints? Yes — posted-content (e.g. arXiv/bioRxiv deposits registered with Crossref) is included; filter with type: posted-content.

Is the full text included? No — Crossref is metadata (and, when the publisher deposits it, an abstract). Use the DOI link for the full text.

Notes & limits

Data from Crossref (api.crossref.org). Citation counts (citedByCount) reflect Crossref's reference index and can differ from other databases. Up to 100 works per call.