mcpField
C
crossref

crossref

Local
github.com/SMABoundlessSearch0Updated 17w ago

About crossref

crossref is an MCP server for searching and retrieving scholarly metadata through the Crossref REST API. It supports work, author, journal, and funder searches, DOI lookups, reference retrieval, and RIS or BibTeX exports without an API key.

Key features

  1. 01Search works by keyword, title, or author
  2. 02Filter and sort search results
  3. 03Look up works by DOI
  4. 04Search journals and funders
  5. 05Retrieve cited references
  6. 06Export results as RIS or BibTeX

Use cases

  1. 01Find papers by topic, author, or year
  2. 02Retrieve metadata for a DOI
  3. 03Gather works and references from a journal
  4. 04Export citations for reference managers

MCP support

It exposes Crossref search, lookup, reference retrieval, and citation export tools to MCP clients. An API key is not required.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"smaboundless-crossref": {
"command": "uvx",
"args": [
"crossref-mcp-server"
],
"env": {
"CROSSREF_MAILTO": "<CROSSREF_MAILTO>"
}
}
}
}
Run commanduvx crossref-mcp-server
Environment variables to set
CROSSREF_MAILTOYour email for Crossref polite pool (better rate limits, no key needed)

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
pypicrossref-mcp-server1.0.0

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

CrossRef MCP Server

An MCP (Model Context Protocol) server for searching and retrieving scholarly metadata from the CrossRef REST API — 150M+ records across all publishers and disciplines.

Built with FastMCP. No API key required.

Tools

ToolDescription
crossref_searchSearch works by keyword with filtering by year, type, and sort options
crossref_title_searchSearch specifically by title for more precise matching
crossref_author_searchSearch for works by a specific author, optionally combined with keywords
crossref_doi_lookupRetrieve full metadata for a work by DOI
crossref_journal_searchSearch for journals by name
crossref_journal_worksGet works published in a specific journal by ISSN
crossref_funder_searchSearch for funding organizations
crossref_referencesGet the reference list cited by a specific work
crossref_export_risExport recent results as RIS (for Zotero, EndNote, etc.)
crossref_export_bibtexExport recent results as BibTeX

Setup

1. Install
cd crossref-mcp-server
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
2. Configure environment (optional)

CrossRef doesn't require an API key, but setting a mailto address gives you access to their faster "polite" API pool:

cp .env.example .env
# Edit .env with your email address
3. Add to Claude Desktop

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "crossref": {
      "command": "/path/to/crossref-mcp-server/venv/bin/python",
      "args": ["/path/to/crossref-mcp-server/server.py"],
      "env": {
        "CROSSREF_MAILTO": "[email protected]"
      }
    }
  }
}

Or if using Claude Code CLI:

claude mcp add crossref \
  /path/to/crossref-mcp-server/venv/bin/python \
  /path/to/crossref-mcp-server/server.py \
  -e [email protected]

Usage examples

Once connected, you can ask Claude things like:

  • "Search CrossRef for recent papers on transformer architectures"
  • "Find works by Jane Smith on educational psychology from 2020-2024"
  • "Look up the metadata for DOI 10.1038/nature14539"
  • "Search for journals about machine learning"
  • "Get the reference list for this paper and export as RIS for Zotero"

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.