Crossref Scraper: 13 Fields, 50K Works/Run, No API Key

Pay only for results delivered. Browse all Actors.

💰 From $0.60 / 1,000 results.

Search Crossref's scholarly record and get 13 fields per work: DOI, title, type, publication year and full date, author list, journal, publisher, citation count, reference count, ISSNs, resolved URL and capture timestamp. Filter by keyword, publication-year range and work type, and pull up to 50,000 works in a single run using Crossref's deep cursor pagination.

Pure HTTP against the official Crossref REST API. No API key, no login, no browser. An optional mailto puts your requests in Crossref's faster polite pool.

Why use this Crossref scraper

• 50,000 works per run, properly paginated. Crossref's offset paging breaks past a few thousand results; this Actor uses the cursor deep-paging API, so a large corpus actually completes.
• Citation counts on every row. cited_by_count (Crossref's is-referenced-by-count) and references_count come back without a second API call, which is what makes bibliometrics possible in one pass.
• Eight work types to filter on. Journal article, book chapter, proceedings article, posted content (preprints), book, dataset and report, or leave it empty for everything.
• A narrow field selection, so responses stay small. The Actor requests only the fields it maps, which keeps large runs fast and cheap.
• Polite pool support. Supply mailto and Crossref routes you to its higher-priority infrastructure at no cost.

Search a topic across a year range

{
  "query": "large language models",
  "fromYear": 2023,
  "toYear": 2026,
  "workType": "journal-article",
  "maxResults": 500
}

The standard literature-review pull: one topic, one window, peer-reviewed articles only.

Build a large bibliometric corpus

{
  "query": "CRISPR gene editing",
  "fromYear": 2015,
  "maxResults": 20000,
  "mailto": "you@example.com"
}

Deep cursor paging plus the polite pool is what makes a 20,000-work pull practical. Leave toYear off for everything published since the start year.

Find conference papers and preprints only

{
  "query": "diffusion models",
  "workType": "proceedings-article",
  "maxResults": 1000
}

Switch workType to posted-content for preprints. In fast-moving fields the preprint record is often months ahead of the journal record.

Track the most-cited work on a topic

{
  "query": "retrieval augmented generation",
  "fromYear": 2022,
  "maxResults": 2000
}

Pull the corpus, then sort locally by cited_by_count, Crossref does not offer a citation-count sort, but the field is on every row so the ranking is a one-liner downstream.

What data does the Crossref scraper return

{
  "doi": "10.1038/s41586-023-06004-9",
  "title": "Example study on language model scaling",
  "type": "journal-article",
  "year": 2023,
  "published_date": "2023-5-17",
  "authors": ["Jane Doe", "John Smith"],
  "journal": "Nature",
  "publisher": "Springer Science and Business Media LLC",
  "cited_by_count": 412,
  "references_count": 63,
  "issn": ["0028-0836", "1476-4687"],
  "url": "https://doi.org/10.1038/s41586-023-06004-9",
  "scraped_at": "2026-08-01T12:00:00.000Z"
}

Field	Description
doi	Digital Object Identifier
title	Work title (first title where a work has several)
type	Crossref work type
year, published_date	Publication year and the fullest date Crossref holds
authors[]	Author names, given plus family, flattened
journal, publisher	Container title and publisher
cited_by_count	How many Crossref-indexed works cite this one
references_count	How many references this work deposits
issn[]	Journal ISSNs
url	Publisher URL, falling back to the DOI resolver
scraped_at	Capture timestamp

Metadata quality on Crossref depends on what each publisher deposits. Missing author lists, dates or ISSNs come back empty rather than reconstructed.

How the scraper works without an API key

Crossref's REST API is open and keyless. This Actor requests 100 works per page with an explicit select field list, follows next-cursor for deep pagination, and retries 429 and 5xx with a capped backoff. Adding mailto opts you into Crossref's polite pool, which is more reliable under load and costs nothing.

What can you build with Crossref data

Literature reviews. A filtered, deduplicated corpus with DOIs, ready to hand to a reference manager or a RAG index.

Bibliometrics and research assessment. Citation and reference counts per work, aggregated by author, journal or year.

Journal and publisher benchmarking. Output volume and citation profile per venue over a period.

Reference resolution. Turn free-text citations into DOIs, then follow them with a citation-graph Actor.

How much does it cost to scrape Crossref

Pay per work delivered: $0.0025 on the Apify Free plan, $0.0015 on Gold and above. A query that matches nothing is never charged.

How do I use Crossref in Claude or ChatGPT

https://mcp.apify.com/?tools=themineworks/crossref-scholarly-metadata

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });

const run = await client.actor('themineworks/crossref-scholarly-metadata').call({
  query: 'large language models',
  fromYear: 2023,
  maxResults: 500,
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();

Crossref scraper FAQ

Do I need an API key? No. Crossref is open. mailto is optional and only improves routing.

What is the polite pool? Crossref runs separate infrastructure for requests that identify themselves. Supplying a real email in mailto gets you onto it, free.

Why is cited_by_count lower than Google Scholar? Crossref counts only citations from works deposited with Crossref. It is a floor, not a total.

How deep can pagination go? Up to 50,000 works per run via cursor paging.

Can I search by author or journal? query is Crossref's free-text search across the record, so an author or journal name works as a query term, though it is not a strict field match.

Complete your research intel pipeline

• OpenCitations Scraper, expand any DOI into its citing and cited works.
• PubMed Scraper, biomedical abstracts and MeSH terms for the same literature.
• arXiv Scraper, the preprint record, usually months ahead of the journal.

Found a bug or want a field added? Open an issue on the Actor's Apify Console page.

Related guides

• Building an Academic Research Data Stack: Crossref, OpenAlex, and Citation-Aware RAG
• Crossref API: 150 Million DOIs, Citation Counts, and Bibliographic Data for Free

Last verified: 2026-08