PubMed Scraper

Extract biomedical articles from PubMed via the NCBI E-utilities API. 37M+ records: PMID, DOI, title, abstract, authors, journal, MeSH terms, keywords, references. No API key, no browser, no proxy.

What does PubMed Scraper do?

Scrape PubMed, the U.S. National Library of Medicine's biomedical literature database, in two modes:

• Search by query — full PubMed syntax: field tags (cancer[Title]), Boolean operators (AND/OR/NOT), quoted phrases, date filters (2024[PDAT]), and publication-type filters. Auto-paginates through results.
• Fetch by PMID — pass exact PubMed IDs, get full records in batch.

Every result includes the complete article metadata: PMID, DOI, PMC ID, title, journal, volume/issue/pages, publication date, author list with affiliations, full abstract (with labeled sections like "BACKGROUND:", "METHODS:"), MeSH descriptors with qualifiers, keywords, publication types, and reference count.

PubMed is fully public. No API key required — the actor sends the NCBI-required tool and email parameters with every request to stay in good standing.

Output fields

Field	Description
pmid	PubMed ID
doi	DOI identifier
pmcId	PubMed Central ID (open-access articles)
title	Article title
journal	Full journal name
journalAbbreviation	ISO journal abbreviation
volume / issue / pages / issn	Bibliographic details
publicationDate	Publication date
authors	Array of { lastName, firstName, initials, fullName, affiliation }
authorCount / firstAuthor / lastAuthor	Author summary
abstract	Full abstract with section labels
abstractLength	Character count
meshTerms	Array of { descriptor, qualifiers[] }
meshCount	Number of MeSH descriptors
keywords	Author-supplied keywords
publicationTypes	Review, Clinical Trial, Meta-Analysis, etc.
referenceCount	Number of cited references
language	Article language
pubmedUrl	Direct PubMed URL

Who is it for?

• Systematic review teams building PRISMA-compliant article datasets
• Medical and life-science researchers sourcing literature for meta-analyses
• AI/ML researchers building biomedical NLP corpora and training sets
• Competitive intelligence in pharma and biotech tracking publication activity
• Medical writers and editors finding evidence for regulatory submissions
• Academic libraries enriching catalogs with abstracts and MeSH data

Pricing

$1.20 per 1,000 results. No subscription.

Results	Cost
100	$0.12
1,000	$1.20
10,000	$12.00

How to use it

1. Pick a Mode.
2. For search: enter a Search Query — PubMed syntax supported.
3. For fetch: enter PMIDs.
4. Optionally filter by Date Range and Publication Types.
5. Optionally set your real Contact Email (required by NCBI).
6. Click Start.

Output example

{
  "recordType": "article",
  "pmid": "38000000",
  "doi": "10.1038/s41586-023-06775-1",
  "pmcId": "PMC10705885",
  "title": "CRISPR-Cas9 genome editing in human primary T cells",
  "journal": "Nature",
  "journalAbbreviation": "Nature",
  "volume": "624",
  "issue": "7992",
  "pages": "415-423",
  "issn": "0028-0836",
  "publicationDate": "2023-12-14",
  "authors": [
    { "lastName": "Smith", "firstName": "Jane", "initials": "J", "fullName": "Jane Smith", "affiliation": "Department of Immunology, MIT" },
    { "lastName": "Chen", "firstName": "Wei", "initials": "W", "fullName": "Wei Chen", "affiliation": "Broad Institute" }
  ],
  "authorCount": 12,
  "firstAuthor": "Jane Smith",
  "lastAuthor": "David Liu",
  "abstract": "BACKGROUND: CRISPR-Cas9 has revolutionized genome editing...\n\nMETHODS: We engineered primary human T cells...\n\nRESULTS: Editing efficiency exceeded 95%...",
  "abstractLength": 1420,
  "meshTerms": [
    { "descriptor": "Gene Editing", "qualifiers": ["methods"] },
    { "descriptor": "CRISPR-Cas Systems", "qualifiers": [] },
    { "descriptor": "T-Lymphocytes", "qualifiers": ["metabolism"] }
  ],
  "meshCount": 14,
  "keywords": ["genome editing", "T cells", "CRISPR"],
  "publicationTypes": ["Journal Article", "Research Support, N.I.H., Extramural"],
  "referenceCount": 48,
  "language": "eng",
  "pubmedUrl": "https://pubmed.ncbi.nlm.nih.gov/38000000/",
  "scrapedAt": "2026-09-25T12:00:00.000Z"
}

Technical details

• Source: NCBI E-utilities API at https://eutils.ncbi.nlm.nih.gov/entrez/eutils/. Free, no key, no login .
• Rate limits: 3 req/s anonymous, 10 req/s with a free NCBI API key. The actor defaults to 400ms between requests and backs off on 429 .
• NCBI policy compliance — every request sends tool= and email= parameters. This is NCBI's documented requirement to avoid IP blocks and to receive early warnings about policy changes.
• esearch → efetch pipeline — esearch returns PMIDs, efetch returns full XML records in batches of 200.
• XML parsing via cheerio in xmlMode — extracts structured records from the PubMed XML schema.
• No browser, no proxy — pure HTTP.

Known limits

• Rate limit is 3 req/s anonymous. For high-volume runs, register a free NCBI API key to raise it to 10 req/s .
• NCBI may block datacenter IPs that overload the servers. The actor's tool and email params satisfy NCBI's registration requirement and prevent this — but running at the rate limit is not advised.
• retmax caps at 10,000 PMIDs per search. For deeper results, paginate with retstart.
• efetch batches cap at ~200 PMIDs per request. The actor handles this automatically.
• Full text is not included. PubMed provides abstracts and metadata; full text lives in PubMed Central (PMC) or publisher sites.
• Retractions and updates are noted in the PubMed record but not surfaced as separate fields.

FAQ

Do I need an API key? No. Anonymous access works at 3 req/s. A free NCBI API key raises the limit to 10 req/s .

Do I need a proxy? No. Datacenter IPs work when sending the required tool and email params.

What query syntax is supported? Full PubMed syntax: field tags ([Title], [Author], [PDAT]), Boolean operators, quoted phrases, and filters. See the PubMed search guide.

How do I search by MeSH term? Use the [MeSH Terms] tag: diabetes[MeSH Terms] AND 2024[PDAT].

How do I search a specific author? Use the [Author] tag: Smith J[Author] AND cancer[Title].

How do I export data? After a run, go to Storage → Export as JSON, CSV, Excel.

Support

Open an issue on the Actor's page for bugs or feature requests.