OpenAlex Research Papers Scraper

📚 OpenAlex Research Papers Scraper

Search and export scholarly papers from OpenAlex, the open catalogue of 300M+ works from every field. Get full abstracts as plain text, DOIs, authors with ORCIDs and institutions, venues, citation counts, topics and open-access PDF links, plus citing works, reference lists, author profiles and institution profiles. No login and no API key needed; 200 results per request, so thousands of papers take seconds.

✨ Why use OpenAlex Research Papers Scraper?

• ⚡ Fast and large: 200 papers per request; up to 10,000 per list without a key, unlimited with your own free OpenAlex key.
• 📝 Real abstracts: OpenAlex stores abstracts as a word index; this actor rebuilds the full abstract as readable text.
• 🔓 Open-access links: OA status, best free link and PDF link for every paper that has one.
• 🔗 Citation graph: papers citing a DOI, the references of a DOI, and papers by an author, in the same run.
• 🎛️ Precise filters: year range, paper type, open access, topic, institution, author, country, language, minimum citations, abstract or PDF available.
• 🤖 Agent-friendly: flat camelCase fields, ids without URL clutter, author names as a simple list; works well as an MCP tool.

🎯 Use cases

• 📖 Literature reviews and meta-analyses: pull every paper on a topic and year range with abstracts for screening.
• 🧠 RAG and AI research assistants: feed abstracts, metadata and PDF links into your retrieval pipeline.
• 📈 Research analytics: compare institutions, authors and topics by output, citations and h-index.
• 🔗 Citation tracking: monitor who cites your papers or a competitor's.
• 🏛️ Grant, hiring and partner scouting: find active researchers on a topic at a given institution or country.
• 📰 Trend monitoring: newest papers on a topic, every week.

🚀 How to use

1. Enter Search papers keywords, or paste DOIs to look up, or DOIs under Papers citing / References of, or names under Papers by author, Author profiles or Institution profiles.
2. Add filters (year range, types, open access, topics, institutions, ...) and a sort.
3. Set Maximum results per list.
4. Click Start and download JSON, CSV, Excel or HTML, or use the API.

📥 Input

{
  "searchQueries": ["retrieval augmented generation"],
  "yearFrom": 2023,
  "openAccessOnly": true,
  "workTypes": ["article", "preprint"],
  "sortBy": "citations",
  "maxResults": 200
}

📤 Output

Paper rows (type: "work"): id, doi, title, publicationYear, publicationDate, workType, language, abstract, authors (name, id, ORCID, position, corresponding, institutions), authorNames, institutions, countries, venue, venueIssn, publisher, volume, issue, firstPage, lastPage, isOpenAccess, oaStatus, oaUrl, pdfUrl, license, landingPageUrl, citedByCount, referencedWorksCount, fwci, citationPercentile, primaryTopic, subfield, field, domain, topics, keywords, concepts, meshTerms, funders, sdgs, isRetracted, pmid, pmcid, countsByYear, optional referencedWorks and relatedWorks, openalexUrl, and sourceType / sourceInput (which input produced the row).

Author rows (type: "author"): id, name, orcid, alternativeNames, worksCount, citedByCount, hIndex, i10Index, twoYearMeanCitedness, lastKnownInstitution(s), countryCode, affiliations (with years), topics, countsByYear.

Institution rows (type: "institution"): id, name, ror, institutionType, countryCode, city, region, country, latitude, longitude, homepageUrl, imageUrl, acronyms, worksCount, citedByCount, hIndex, i10Index, parentInstitutions, topics, countsByYear.

{
  "type": "work",
  "sourceType": "lookup",
  "sourceInput": "https://arxiv.org/abs/1706.03762",
  "id": "W2626778328",
  "title": "Attention Is All You Need",
  "publicationYear": 2025,
  "workType": "preprint",
  "language": "en",
  "abstract": "The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. …",
  "authors": [{"name": "Jane Example", "id": "A5000000001", "orcid": null, "position": "first", "isCorresponding": false, "institutions": ["Example University"]}],
  "authorNames": ["Jane Example", "John Sample"],
  "isOpenAccess": true,
  "oaStatus": "gold",
  "pdfUrl": "https://example.org/preprint.pdf",
  "citedByCount": 24501,
  "referencedWorksCount": 28,
  "primaryTopic": "Natural Language Processing Techniques",
  "field": "Computer Science",
  "keywords": ["transformer models", "attention mechanism", "machine translation"],
  "openalexUrl": "https://openalex.org/W2626778328"
}

💰 Pricing

• $3.20 per 1,000 results (papers, authors or institutions).
• Example: 1,000 papers with abstracts ≈ $3.20 in result fees plus well under a cent of platform usage.
• Failed lookups and empty searches are never charged.

❓ FAQ

Is this legal? Yes. OpenAlex is an open catalogue; its data is released under CC0. The actor uses the public OpenAlex API.

Do I need an API key? No. The actor works without one. For very large jobs you can add your own free OpenAlex key in Advanced.

How many results can I get? Up to 10,000 per list (search, citing list, filter set) without an API key, which is OpenAlex's own paging limit. Add your free OpenAlex key in Advanced to page through everything, or split a big list by year.

Why is the abstract empty for some papers? OpenAlex has no abstract for those papers (common for older works and some publishers). Use Only papers with an abstract to skip them.

How do I get the papers citing a DOI? Put the DOI in Papers citing. Filters and sort apply to that list too.

Can I call it from an AI agent? Yes: the inputs are simple strings and numbers, every row has clear field names, and it runs as a tool over MCP or the API.

⚠️ Limits

• Without an API key OpenAlex pages at most 10,000 results per list; with your free key there is no limit.
• Coverage and metadata quality are OpenAlex's: some papers lack abstracts, venues or affiliations.
• Name inputs (author, institution, topic) use OpenAlex's best match; use ids, ORCIDs or ROR ids for exact control.
• Full text is not downloaded; pdfUrl and oaUrl point to where it can be read.