MCP.so
Sign In
Servers

OpenAlex MCP Server

@reetp14

openalex-mcp is a mcp server which provide AI agents with tools to interact with OpenAlex.

Overview

What is OpenAlex MCP Server?

A Model Context Protocol (MCP) server that provides access to the OpenAlex API – a fully open catalog of the global research system covering over 240 million scholarly works. It's for developers and researchers who want to search and retrieve scholarly works, authors, sources, institutions, concepts, publishers, and funders.

How to use OpenAlex MCP Server?

Install via npm install -g openalex-mcp and add the server to your MCP client configuration. Use tools such as search_works, search_authors, search_sources, search_institutions, search_concepts, search_publishers, search_funders, get_entity, autocomplete, and classify_text with parameters like search, filter, sort, page, cursor, group_by, select, sample, and mailto.

Key features of OpenAlex MCP Server

  • Search over 240 million scholarly works
  • Retrieve authors with ORCID integration
  • Access ~250K sources (journals, conferences, repositories)
  • Institutions with ROR matching
  • Hierarchical research concepts (levels 0-5)
  • Full OpenAlex query grammar for filtering and sorting

Use cases of OpenAlex MCP Server

  • Search for AI papers from 2023 and sort by citation count
  • Find authors affiliated with a specific institution
  • Get publication trends by year using grouping
  • Autocomplete journal names for user input
  • Classify research text to predict relevant concepts

FAQ from OpenAlex MCP Server

What is the OpenAlex API?

It is a fully open catalog of the global research system, free and open, created by OurResearch.

What are the rate limits?

Anonymous: 10 requests/second, 100,000/day. With mailto or Bearer token: 100 requests/second, 1,000,000/day.

How do I authenticate?

Optionally use a Bearer token via the OPENALEX_BEARER_TOKEN environment variable or provide a mailto parameter in requests.

Can I paginate deeply?

Yes, use the cursor parameter with * for the first call, then use the returned next_cursor. Standard pagination max 10,000 results; deep pagination allows more.

What response format does it return?

Standard OpenAlex JSON envelope with meta (count, db_response_time_ms, page, per_page, next_cursor) and results array.

More from Other