Academic Research Search (OpenAlex)

Search over 250 million scholarly works from OpenAlex by topic, author, or institution, with citations, authors, venue, and open access status.

What it does

• Full-text search across titles, abstracts, and body text using the OpenAlex works API.
• Filter by publication year, open access status, and minimum citation count.
• Returns clean, flat paper records: title, DOI, year, citation count, authors, institutions, venue, open access status and link, topics, and type.
• Exports an aggregate summary (total matched, returned, most cited) to OUTPUT.
• No API key needed.

Example input

Recent, well-cited papers on a topic:

{
  "search": "large language models",
  "fromYear": 2023,
  "minCitations": 10,
  "maxItems": 40
}

Open access papers only:

{
  "search": "CRISPR gene editing",
  "openAccessOnly": true,
  "maxItems": 50
}

Input

Field	Type	Description
search	string	Full-text search term: a topic, author name, or institution. Required.
fromYear	integer	Only keep works published in or after this year. Leave empty for no lower bound.
openAccessOnly	boolean	Only keep works that are freely readable in some open access location. Default false.
minCitations	integer	Only keep works cited at least this many times. Default 0.
maxItems	integer	How many works to return (auto-paginated, max 10000). Default 40.

Output

Each dataset item is one work:

{
  "title": "ChatGPT for good? On opportunities and challenges of large language models for education",
  "doi": "https://doi.org/10.1016/j.lindif.2023.102274",
  "year": 2023,
  "citedByCount": 5259,
  "authors": ["Enkelejda Kasneci", "Kathrin Seßler"],
  "institutions": ["Technical University of Munich"],
  "venue": "Learning and Individual Differences",
  "isOpenAccess": true,
  "oaUrl": "https://epub.ub.uni-muenchen.de/125071/1/ChatGPT_for_Good_v3.pdf",
  "topics": ["Artificial Intelligence in Healthcare and Education", "Topic Modeling"],
  "type": "article",
  "openAlexId": "https://openalex.org/W4323655724"
}

Run summary (OUTPUT)

The run's default key-value store record OUTPUT holds an aggregate for the whole result set:

{
  "query": { "search": "large language models", "fromYear": 2023, "openAccessOnly": false, "minCitations": 10 },
  "totalMatched": 184213,
  "returned": 40,
  "requestedMaxItems": 40,
  "scanned": 40,
  "mostCited": {
    "title": "...",
    "citedByCount": 5259,
    "openAlexId": "https://openalex.org/W4323655724",
    "doi": "https://doi.org/10.1016/j.lindif.2023.102274"
  },
  "generatedFrom": "https://api.openalex.org/works"
}

How it works

The Actor calls the public OpenAlex works API with your search term and paginates until it has maxItems results or runs out of matches. Year, open access, and citation filters are combined into a single OpenAlex filter expression and applied server-side. Every request includes a mailto parameter for the polite pool and a descriptive User-Agent, and retries on rate limits (429) and server errors with backoff, honoring the API's Retry-After header.

Use cases

• Literature review: pull recent, well-cited papers on a research topic.
• Track output from an author or institution over time.
• Find open access versions of papers for a reading list.
• Feed structured citation data into research dashboards or agents.

MIT licensed.