OpenAlex Research Intelligence

Extract research intelligence from the official OpenAlex API without browser scraping. Use search terms, OpenAlex work IDs, author IDs, institution IDs, or concept IDs to produce flattened dataset rows for works, authors, and publication sources.

Run the next report

• Turn a literature watch into a source-linked trial and evidence-gap report with Biomedical Trial & Literature Evidence Report.

Use cases

Use this actor for academic / bibliometric research — query 250M+ scholarly works by topic, author, institution, or citation. Auth-free, official-API-first, with a stable output schema and documented source compliance.

Monitor mode - new-paper alerts (recurring)

Schedule this actor and get only the works you haven't seen before on each run - a stateful "new research" alert, not a full re-dump. Ideal for watching an author, institution, or concept for new publications.

• Set monitor: true and run on a schedule (Apify Schedules).
• The first run defaults to initialRunMode: "baseline_only": it establishes state, emits 0 rows, and makes 0 PPE charges.
• Use initialRunMode: "emit_backfill" when you explicitly want the first baseline delivered as charged rows for a billing proof or initial export.
• Every later run emits only new works (plus their author/source rows). If the state is unchanged, it emits 0 rows and makes 0 charges.
• State is namespaced per query — watching several? Give each a distinct monitorKey.
• A state read/save failure fails closed: the actor does not charge rows or advance the cursor.

{
  "institutionIds": ["I63966007"],
  "sort": "publication_date:desc",
  "monitor": true,
  "initialRunMode": "baseline_only",
  "delivery": "webhook",
  "webhookUrl": "https://your-endpoint.example/new-papers"
}

Tip: entity queries (authorIds / institutionIds / conceptIds) are the most reliable for scheduled monitoring.

Inputs

Field	Default	Notes
searchTerms	[]	Research queries for OpenAlex works.
workIds	[]	OpenAlex work IDs or URLs.
authorIds	[]	OpenAlex author IDs.
institutionIds	[]	OpenAlex institution IDs.
conceptIds	[]	OpenAlex concept IDs.
fromDate / toDate	empty	Publication date filters in YYYY-MM-DD.
sort	cited_by_count:desc	Citation, publication date, or relevance sort.
limitPerSource	25	Works fetched per search/filter source.
maxWorks	100	Global unique work cap.
includeAbstract	false	Reconstruct abstract text from OpenAlex inverted index when available.
mailto	empty	Optional polite-pool email parameter for OpenAlex.
monitor	false	Monitor mode: emit only works not seen on a prior run (stateful new-results alert). Pair with a schedule.
monitorKey	empty	Optional label to namespace monitor state across several watched queries.
initialRunMode	baseline_only	First monitor run: baseline_only records state with 0 rows/charges; emit_backfill delivers the current baseline as charged rows.
delivery	dataset	dataset or webhook.
dryRun	false	Skip dataset/webhook delivery.

At least one of searchTerms, workIds, authorIds, institutionIds, or conceptIds is required.

Dataset Rows

work_summary

• title, DOI, OpenAlex ID, publication year/date, type
• citation count, open-access status, retraction flag
• primary source, publisher, landing page, PDF URL
• topics, concepts, countries, institutions

author_signal

• work ID/title, author name and OpenAlex ID
• author order/position, corresponding author flag
• affiliated institutions and countries

source_summary

• work ID/title, journal or source name
• source type, ISSN, host organization, open-access source flag

Example Input

{
  "searchTerms": ["large language models", "retrieval augmented generation"],
  "fromDate": "2024-01-01",
  "sort": "cited_by_count:desc",
  "limitPerSource": 10,
  "maxWorks": 20,
  "includeAbstract": false,
  "delivery": "dataset",
  "dryRun": false
}

Sample output

Each run produces structured dataset rows (see the Dataset Rows section above for the field list). Run the actor once with the example input to see a live sample before scheduling.

Local Development

npm install
npm test
node src/index.js

output/result.json contains the full payload. Apify dataset delivery writes flattened rows.

Limitations

• OpenAlex coverage is broad but not identical to Crossref, PubMed, Semantic Scholar, or publisher APIs.
• Citation counts and metadata can lag source publications.
• Relevance sort is used only for search sources; non-search filters fall back to citation sort.
• includeAbstract can increase payload size substantially.

Monitoring example

See Monitor mode above to schedule new-paper alerts for an author, institution, or concept.

Billing safety

• monitor: false preserves one-shot behavior and delivers each output row through the configured PAY_PER_EVENT event.
• monitor: true persists the seen-work cursor only after delivery and the PHASE89_DELIVERY_AUDIT receipt succeed.
• The runtime verifies the chargedCount returned by Actor.pushData for every billable row. Missing charge receipts fail closed.