Log inSubmit server
Openaire logo

Openaire

Searches public OpenAIRE records for EU-funded publications, datasets, projects, and software through MCP.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Openaire

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "openaire": {
      "command": "npx",
      "args": [
        "-y",
        "@pipeworx/mcp-openaire"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

We couldn’t confirm it starts correctly

The install command started, but didn't respond the way we expected when we tried to talk to it.

npx -y @pipeworx/mcp-openaire

initialize succeeded but no response to tools/list.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 16d ago.

Overview

Openaire MCP server connects AI agents to the public OpenAIRE Graph API for EU research publications, datasets, projects, and software. It provides six search and record-retrieval tools over a remote gateway or as a local stdio package. Use it to find research outputs by topic and, where supported, filter publications and projects by funder, country, or year. No authentication is required.

Use cases

  • Search publications on a research topic
  • Filter funded projects by country or year
  • Find datasets related to a query
  • Retrieve a publication or project by ID
  • Search research software records

Key features

  • Search OpenAIRE publications, datasets, projects, and software
  • Filter publication and project searches by funder, country, or year
  • Retrieve publications and projects by ID
  • Paginate supported search results
  • Run through hosted MCP or local stdio

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 3d ago

Recent health check succeeded.

7/7 checks healthy over the last 51d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 3d ago via OSV.dev · @pipeworx/mcp-openaire (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

npm downloads485Package downloads in the last 30 days.
Last commit16d agoMost recent push to the default branch.
Install checkInconclusiveStarted but did not complete the MCP handshake in our test.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How Openaire works

What Openaire MCP server does

The Openaire MCP server exposes selected OpenAIRE Graph API operations through Model Context Protocol. It covers four types of research output: publications, datasets, projects, and software. The underlying data source is the public API at https://api.openaire.eu/graph/v1/, which returns JSON.

Search tools accept a text query. Publication and project searches can also use optional funder, country, and year filters. Dataset, publication, project, and software searches support pagination controls through size and page, where available. Separate retrieval tools fetch an individual publication or project by ID.

How it works

You can connect an MCP client to the hosted endpoint at https://gateway.pipeworx.io/openaire/mcp. That endpoint lists the Openaire tools together with shared Pipeworx gateway tools, including routing and discovery functions. Its exact tool scope is reported by the MCP initialize response and may change, so clients should treat that response as authoritative.

The Openaire MCP server can also run locally through the @pipeworx/mcp-openaire npm package. The local process communicates over stdin and stdout and exposes only the pack-specific tools. It does not add the shared gateway meta-tools or provide cross-catalog routing.

For clients that do not use MCP, the hosted gateway also exposes HTTP tool endpoints. For example, publication search is available through a POST request to /v1/tools/openaire_search_publications, with JSON arguments in the request body.

Setup and configuration

For a hosted MCP connection, add the gateway URL as a remote server in an MCP client. The documented examples name Claude Desktop, Cursor, and Windsurf. No account or authentication is required for the initial calls.

For a local connection, configure the client to run npx -y @pipeworx/mcp-openaire as the MCP command. This option avoids the gateway round trip and does not require a Pipeworx account. No environment variables are documented as required.

A full Pipeworx gateway endpoint is also available at https://gateway.pipeworx.io/mcp. It lists the broader catalog directly, while the Openaire endpoint lists this pack directly and still includes shared gateway tools.

Tools and capabilities

The Openaire MCP server provides these pack-specific tools:

  • search_publications(query, funder?, country?, year?, size?, page?)
  • search_datasets(query, size?, page?)
  • search_projects(query, funder?, country?, year?, size?, page?)
  • search_software(query, size?, page?)
  • get_publication(id)
  • get_project(id)

Documented funder codes include European Commission (ec__________::EC), NIH (nih_________::NIH), Wellcome Trust (wt__________::WT), and NSF (nsf_________::NSF).

Limitations and notes

The hosted pack endpoint includes shared Pipeworx meta-tools in addition to the six Openaire tools, which increases the tool count shown to a client. There is no documented way to remove those shared tools from a hosted pack connection. Use the local stdio package when only the Openaire tool set is needed.

The material documents filtering by funder, country, and year only for publications and projects. It does not describe equivalent filters for datasets or software.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Openaire

Openaire is an MCP server that connects AI agents to the public OpenAIRE Graph API for EU research outputs. Its main tools search publications, datasets, projects, and software, and retrieve publications or projects by ID. It supports hosted MCP access through Pipeworx or local stdio execution without authentication.

Install it locally by configuring your MCP client to run `npx -y @pipeworx/mcp-openaire`. The package communicates over stdin and stdout, needs no Pipeworx account, and requires no documented environment variables.

Yes, Openaire works with Cursor through the hosted MCP endpoint. Add `https://gateway.pipeworx.io/openaire/mcp` as an MCP server in the client configuration.

No, Openaire does not require authentication for the documented hosted calls or local package. The hosted service uses the public OpenAIRE API, and the standalone stdio mode does not require a Pipeworx account.

Call `search_publications` with a query and optionally provide funder, country, year, size, or page arguments. The hosted gateway also accepts HTTP POST requests at `/v1/tools/openaire_search_publications`.

Related MCP servers

All in Research All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewOpenaire AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/openaire?style=directory)](https://allmcps.com/mcp/openaire)
HTML Embed
<a href="https://allmcps.com/mcp/openaire"><img src="https://allmcps.com/api/badge/openaire?style=directory" alt="Openaire on AllMCPs" /></a>