Crossref Works Search

Crossref API without a key or code: search more than 180 million scholarly works by keyword, author, journal, funder or DOI and get clean rows.

What this Crossref API Actor does

Crossref Works Search looks through more than 180 million scholarly works registered with Crossref: journal articles, book chapters, conference papers, preprints, datasets, reports and more. Search by keywords, author, journal name, ISSN, publication date, work type and funder, or paste a list of DOIs to look them up in bulk. You get one row per work with the DOI, title, author names, journal, publisher, publication dates, volume, issue and pages, citation count, number of references, license, funders and full-text links. Abstracts are optional: turn on Include abstracts to add the abstract text when the publisher deposited one.

Use it for literature reviews, citation counts for a list of papers, a journal's output for a year, everything a funder paid for, or a clean DOI metadata export for your reference manager, spreadsheet or AI pipeline. The data comes from Crossref's public metadata, read live when you run it. No login, no API key and no Crossref account are needed.

How to search Crossref works, step by step

1. Open the Actor and type Search words, for example large language models. Or leave it empty and use Author, Journal name, Journal ISSNs or Funders instead.
2. Narrow the results with Published from and Published until (2024, 2024-05 or 2024-05-17) and Work types (journal article, book chapter, conference paper, preprint and others).
3. Choose Sort by: relevance, newest, oldest, most cited or recently updated.
4. Set Maximum results (1 to 10,000) to cap the run and the cost.
5. To look up known papers instead, paste them into DOIs (plain DOIs or doi.org links, up to 10,000). The search fields are not used when DOIs are given.
6. If you need abstracts, open Advanced options and turn on Include abstracts. Abstracts are off by default.
7. Click Start. Results appear in the Overview table and can be downloaded as JSON, CSV, Excel or HTML, or read through the Apify API.

How much does a Crossref works search cost?

Pay per event: you pay $0.75 per 1,000 works ($0.00075 per work) on the Free plan, $0.675 on Starter, $0.60 on Scale and $0.525 on Business, plus the standard Actor start event. Only rows with work data are charged. DOIs that Crossref does not know (status not_found) and entries that are not DOIs (status invalid_input) return a free row.

Worked example (Free plan prices): a literature scan of 2,000 journal articles costs 2,000 x $0.00075 = $1.50. Citation counts for a reading list of 150 DOIs cost about $0.11. A weekly run of the 500 newest papers on a topic costs about $0.38 per run. Set Published from to the date of the previous run, since a work returned in an earlier run is returned and charged again. Set a spending limit on the run and the Actor stops cleanly when the limit is reached.

Input example for a Crossref works search

{
    "query": "large language models",
    "publishedFrom": "2025-01",
    "types": ["journal-article"],
    "sortBy": "most-cited",
    "maxResults": 500
}

DOI lookup:

{
    "dois": ["10.1038/nature12373", "https://doi.org/10.1126/science.1225829"]
}

Field	Console name	What it does
query	Search words	Words to find in titles, abstracts, authors, journals and other metadata
author	Author	Author name to match
journal	Journal name	Journal, book series or proceedings title to match
issns	Journal ISSNs	Works from these journals only
publishedFrom	Published from	Earliest publication date
publishedUntil	Published until	Latest publication date
types	Work types	journal-article, book-chapter, proceedings-article, posted-content and more
funders	Funders	Funder Registry IDs or exact funder names
dois	DOIs	Look up these works directly
sortBy	Sort by	relevance, newest, oldest, most-cited, recently-updated
maxResults	Maximum results	1 to 10,000 works from a search
onlyWithAbstract	Works with an abstract only	Skip works without an abstract
includeAbstracts	Include abstracts	Add the abstract text to each row (off by default)

Output example: one Crossref work

{
    "doi": "10.1038/nature12373",
    "doiUrl": "https://doi.org/10.1038/nature12373",
    "title": "Nanometre-scale thermometry in a living cell",
    "authors": ["G. Kucsko", "P. C. Maurer", "N. Y. Yao", "M. Kubo", "H. J. Noh", "P. K. Lo", "H. Park", "M. D. Lukin"],
    "authorCount": 8,
    "journal": "Nature",
    "issn": ["0028-0836", "1476-4687"],
    "publisher": "Springer Science and Business Media LLC",
    "type": "journal-article",
    "publishedDate": "2013-07-31",
    "publishedYear": 2013,
    "volume": "500",
    "issue": "7460",
    "pages": "54-58",
    "citationCount": 1827,
    "referencesCount": 30,
    "license": "http://www.springer.com/tdm",
    "hasAbstract": false,
    "abstract": null,
    "funders": [],
    "publisherUrl": "https://www.nature.com/articles/nature12373",
    "status": "ok"
}

Other fields in every row: subtitle, journalAbbreviation, isbn, publishedPrintDate, publishedOnlineDate, licenses, subjects, links, relevanceScore, createdDate, indexedAt and error. A STATS record in the key-value store gives the number of matching works, the works delivered and the charges for the run.

Data fields in the Crossref works dataset

• doi, doiUrl: the DOI in lower case and its doi.org link.
• title, authors, authorCount: the work's title and author names as published. Only names are returned.
• journal, issn, publisher, type: where and by whom the work was published.
• publishedDate, publishedYear: the earliest publication date Crossref holds, print or online.
• citationCount: how many works registered with Crossref cite this one. Other citation indexes count other sources, so their numbers differ.
• referencesCount: how many references the work cites.
• license: the license URL for the version of record, or the first license listed. Some publishers list a text-and-data-mining license here, not an open license.
• hasAbstract, abstract: whether the publisher deposited an abstract (about a quarter of works), and the abstract as plain text. The text is filled only when you turn on Include abstracts; otherwise it is null.
• funders: funder names with their Funder Registry ID and award numbers.

FAQ

Is it legal to use Crossref metadata?

Crossref states that almost none of its metadata is subject to copyright and that it may be used for any purpose. Abstracts are the exception: they can remain the publisher's copyright. That is why abstracts are off by default. Turn on Include abstracts only if you need them. This Actor reads only public metadata, needs no login and returns author names only, with no contact details.

Can I reuse the abstracts?

Abstracts are the publisher's text, and you are responsible for how you reuse them.

Which works are included?

Works with a Crossref DOI: most journals, many books and conference proceedings, and preprint servers that register DOIs. Works registered only with another DOI agency are not included. The Actor reads Crossref's open metadata, so there are no CAPTCHAs or blocks.

How many results can I get?

Up to 10,000 works per run from a search, and up to 10,000 DOIs per run in a DOI lookup. For a larger set, split it by year or work type and run once per part.

Why do my search results include loosely related works?

Search words match any of the words, ranked by relevance. Sort by relevance and use a small Maximum results, or add Work types, dates or a journal to narrow it down.

Why does a DOI return not_found?

Crossref has no record for it. The DOI may be registered with another DOI agency or may contain a typing error. These rows are free.

Why is a funder name rejected?

Funder names must match the Funder Registry name exactly. The error message lists the closest matches with their IDs; paste the ID into Funders.

Is this Actor affiliated with Crossref?

No. It is an independent tool that reads Crossref's public metadata. Crossref is a trademark of its owner.

Run it every day

Use this to get each new work once.

1. Enter your search words, filters or DOIs.
2. Turn on Only new since the last run and give the monitor a Monitor name.
3. Click Save as a new task.
4. Open Schedules, create a schedule, pick the task and choose daily.

The first run returns the works that match and sets the baseline. Each later run returns only works that this monitor did not deliver before. Works delivered before are not charged. For a search, the Actor looks at up to the first 50,000 matches. Set Sort by to Recently updated to see the latest changes first; the run then stops at the first page that holds only works delivered before. The memory is in your own Apify key-value store named crossref-monitor- followed by the Monitor name. It holds DOIs only. Use one Monitor name for each task.

Example: 40 new works a day cost 40 x $0.00075 = $0.03 a day (the 40 works are an example, not a measured value). A run without new works costs only the small Apify run start fee.

Bulk lists

Use a long list of DOIs from an Apify Dataset or from a file. Pick the Dataset under Dataset with the DOIs and write the name of its field under Field in the dataset. Leave the field name empty to use the first text field. Or paste the web address of a .txt, .csv or .json file under DOI file URL. A .csv file uses its first column. A header row is skipped. The file can be up to 10 MB. Entries from the Dataset and the file are added to the ones typed under DOIs, and duplicates are removed. The price per result does not change.

Related Actors

• Article Extractor: Use it to pull clean article text, dates and metadata from news and blog pages.
• EU & UK Public Tenders API: Use it to search public tender notices from TED (EU) and Find a Tender (UK).
• H-1B Visa Sponsorship Salary Data: Use it to check offered wages and prevailing wages from US Department of Labor H-1B and PERM files.
• Web Page to Markdown for AI: Use it to turn web pages into clean Markdown for LLMs, RAG and AI agents.