canonical: https://jentic.com/apis/openalex.org/openalex

# OpenAlex API

OpenAlex is a free, open catalog of the global research system covering scholarly works, authors, sources, institutions, topics, publishers, and funders. The REST API exposes 16 endpoints that let agents query and retrieve detailed metadata on more than 250 million scholarly works and their citation graph. Responses are returned as JSON and the API supports filtering, search, grouping, and cursor-based pagination across each entity type. An optional API key (passed as a query parameter) raises rate limits and grants access to the polite pool.

## For AI agents

Search, retrieve, and analyze scholarly works, authors, institutions, sources, and topics from the OpenAlex catalog through 16 read-only endpoints.

## Scope

Does not handle full-text PDF downloads, peer-review submissions, or DOI registration - use for scholarly metadata search and citation graph traversal only.

## Capabilities

- Search scholarly works by title, abstract, DOI, or full-text query and filter results by author, year, or institution
- Retrieve a complete author profile including affiliations, h-index, citation counts, and concept tags
- Look up an institution by ROR ID and pull its associated works, authors, and country metadata
- Group works by publication year, host source, or topic to build citation analytics dashboards
- Fetch a single work by OpenAlex ID, DOI, or PMID and extract its referenced and citing works
- Discover sources (journals, repositories, conferences) ranked by impact and open-access status
- Traverse the topic and concept hierarchy to map research areas across millions of papers

## Use cases

### Literature Review Automation

Build a literature review pipeline that pulls every relevant scholarly work for a research topic, enriches each with author and institution metadata, and exports a structured bibliography. OpenAlex covers more than 250 million works and is free to query, making it suitable for systematic reviews, grant proposals, and competitive intelligence projects without per-call licensing costs. The API supports cursor pagination so an agent can walk the full result set for a query in minutes.

Example prompt: Search /works for the query 'large language model evaluation' filtered to publication_year:2024-2025, page through results with cursor pagination, and return a deduplicated list of titles, DOIs, and author names.

### Researcher Profile Lookup

Resolve a researcher's full profile from a name or ORCID, returning their works, affiliations, citation totals, and topic clusters. This supports talent search, conflict-of-interest screening, and grant reviewer recommendation. Each author endpoint returns counts, concepts, and a works URL that an agent can follow to retrieve their full publication list.

Example prompt: Call GET /authors with a search filter on display_name='Yann LeCun', then retrieve the top match's works_api_url and return the 10 most-cited publications.

### Institution Research Analytics

Aggregate research output by institution to build dashboards showing publication counts, topic distribution, and citation impact. OpenAlex links works to ROR-identified institutions, so an agent can compare two universities or track department output over time. Group-by parameters return faceted counts in a single request without paging through every record.

Example prompt: Call GET /works with filter institutions.ror=https://ror.org/02jx3x895 and group_by=publication_year, then return yearly publication counts for the last decade.

### Citation Graph Traversal

Walk the citation graph by following referenced_works and cited_by_api_url fields on each work. Agents can build seed-and-expand discovery tools that start from a single paper and surface related literature within a few hops. The graph covers all indexed works, providing free coverage similar to Web of Science or Scopus without subscription gates.

Example prompt: Fetch a work by DOI via GET /works/doi:10.1038/nature14539, then retrieve all entries from its cited_by_api_url and return titles plus citation counts.

### Agent-Driven Research Assistant via Jentic

Wire OpenAlex into an AI research assistant through Jentic so an agent can answer scholarly questions with cited evidence. The agent uses Jentic intent search to discover the right OpenAlex operation, loads the request schema, and executes the call with the user-supplied API key isolated in your Jentic One instance. Because OpenAlex offers free access without OAuth, integration through Jentic typically lands inside an hour.

Example prompt: Use Jentic search query 'find scholarly works on a topic', load the OpenAlex /works schema, and execute a search for 'graph neural networks' returning the 5 most-cited 2024 papers.

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | /works | Search and filter scholarly works |
| GET | /works/{id} | Retrieve a single work by OpenAlex ID or DOI |
| GET | /authors | Search authors by name, ORCID, or filter |
| GET | /authors/{id} | Retrieve a single author profile |
| GET | /sources | Search journals, conferences, and repositories |
| GET | /institutions | Search institutions by ROR ID or country |
| GET | /topics | Browse research topics and subfields |

## Key resources

- **Works** — Search, retrieve, filter, and group the 250M+ scholarly works (papers, books, datasets) in the catalog.
- **Authors** — Retrieve author profiles with affiliations, citation counts, and works URLs for further traversal.
- **Sources** — Look up journals, conferences, and repositories with impact metrics and open-access flags.
- **Institutions** — Resolve universities and labs by ROR ID with country, type, and aggregated output metadata.
- **Topics** — Browse the topic hierarchy that classifies works into research areas and subfields.

## Why Jentic

- **Setup:** Wiring the OpenAlex API by hand means passing your api_key query parameter, learning its filter syntax, and paging the works and authors results yourself. Through Jentic you install once, import the OpenAlex API from the API Directory, store the API key once, and your agent calls it.
- **Permission scoping:** OpenAlex puts the entity id in the URL path (/works/{id}, /authors/{id}), so a rule can pin your agent to reading a specific work or author. You choose the operations it may call, and since the API is read-only the agent gets only the lookups you allow.
- **Credential handling:** Your OpenAlex API key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'find scholarly works citing a paper' or 'look up an author by ORCID', and Jentic returns the matching OpenAlex operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **CORE API** — Aggregator of open-access research papers with full-text search and download.
- **Crossref REST API** — DOI registration agency providing canonical scholarly metadata and references.

## FAQ

### What authentication does the OpenAlex API use?

OpenAlex uses an optional API key passed as the api_key query parameter. Anonymous access is allowed but rate-limited; supplying a key places requests in the polite pool with higher limits. When called through Jentic the key is stored in the encrypted vault and injected at execution time, so the raw value never enters agent context.

### Can I retrieve a paper by its DOI through the OpenAlex API?

Yes. Call GET /works/{id} where {id} is the DOI prefixed with 'doi:' (for example /works/doi:10.1038/nature14539). The response contains the full work record including authors, citation counts, referenced works, and the cited_by_api_url for forward citation traversal.

### What are the rate limits for the OpenAlex API?

OpenAlex publishes a default of 100,000 requests per day and 10 requests per second for users in the polite pool (those who supply an api_key query parameter or include their email in the User-Agent header). Anonymous traffic is limited more aggressively. Always page with cursor rather than offset for large result sets.

### How do I search for works on a topic with OpenAlex through Jentic?

Run the Jentic search query 'find scholarly works on a topic', load the schema for GET /works, and execute with parameters such as search='your topic' and filter='publication_year:2024'. Jentic returns the structured response so the agent can read titles, authors, and citation counts directly.

### Is the OpenAlex API free?

Yes, OpenAlex is free to use under a CC0 data license and does not require payment for any tier. Supplying an email or api_key only changes which rate-limit pool the request lands in; it does not unlock paid features.

### How do I traverse citations from one paper to its references with OpenAlex?

Each work record returned by GET /works/{id} includes a referenced_works array (backward citations) and a cited_by_api_url (forward citations). Fetch the cited_by_api_url to retrieve the full list of works that cite the seed paper, paged via cursor.

### Can I limit what my agent is allowed to do with the OpenAlex API?

Yes. Because you run Jentic One yourself, your own rules decide which OpenAlex operations the agent may call, so you can allow only the reads you want, such as GET /works or GET /authors, and block the rest. Since OpenAlex puts the entity id in the URL path (/works/{id}, /authors/{id}), a rule can even pin the agent to a specific work or author. The API is read-only, so the agent only ever gets the lookups you permit, and your API key stays under your control and is injected at execution time rather than exposed to the agent.