mcpField
C
Crossref MCP Server

Crossref MCP Server

Local
github.com/botanicastudiosSearch4Updated 83w ago

About Crossref MCP Server

Crossref MCP Server provides access to the Crossref API for clients that need scholarly work metadata. It supports title and author searches as well as DOI-based work lookups, returning structured JSON.

Key features

  1. 01Search works by title
  2. 02Search works by author
  3. 03Retrieve work details by DOI
  4. 04Return structured JSON responses

Use cases

  1. 01Find publications by title keywords
  2. 02Locate works associated with an author
  3. 03Retrieve metadata for a known DOI

MCP support

Exposes Crossref title search, author search, and DOI lookup as tools for MCP clients.

Configuration

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

mcp.json · Universal config
{
"mcpServers": {
"crossref": {
"command": "npx",
"args": [
"-y",
"@botanicastudios/crossref-mcp"
]
}
}
}
Run commandnpx -y @botanicastudios/crossref-mcp

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
npm@botanicastudios/crossref-mcp

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

A Model Context Protocol (MCP) server for interacting with the Crossref API.

Features

  • Search works by title
  • Search works by author
  • Get work details by DOI

Installation

{
  "mcpServers": {
    "crossref": {
      "command": "npx",
      "args": [
        "-y",
        "@botanicastudios/crossref-mcp"
      ]
    }
  }
}

Usage

The server provides three main tools:

1. Search by Title

Search for works in Crossref by title:

// Example: Search for works containing "quantum computing" in the title
{
  "title": "quantum computing",
  "rows": 5  // Optional, defaults to 5
}
2. Search by Author

Search for works in Crossref by author:

// Example: Search for works by "Einstein"
{
  "author": "Einstein",
  "rows": 5  // Optional, defaults to 5
}
3. Get Work by DOI

Retrieve a specific work using its DOI:

// Example: Get work with DOI "10.1088/1742-6596/1398/1/012023"
{
  "doi": "10.1088/1742-6596/1398/1/012023"
}

Response Format

All responses are returned as structured JSON objects with the following format:

For successful searches:
{
  "status": "success",
  "query": {
    /* the original query parameters */
  },
  "count": 5,
  "results": [
    {
      "title": "Work title",
      "authors": [
        {
          "given": "First name",
          "family": "Last name",
          "name": "First name Last name"
        }
      ],
      "published": {
        "dateParts": [2023, 1, 15],
        "dateString": "2023-1-15"
      },
      "type": "journal-article",
      "doi": "10.xxxx/xxxxx",
      "url": "https://doi.org/10.xxxx/xxxxx",
      "container": "Journal Name",
      "publisher": "Publisher Name",
      "issue": "1",
      "volume": "42",
      "abstract": "This is the abstract of the work, if available."
    }
    // additional results...
  ]
}
For single DOI lookup:
{
  "status": "success",
  "query": { "doi": "10.xxxx/xxxxx" },
  "result": {
    // work details as shown above
  }
}
For errors or no results:
{
  "status": "error" | "no_results" | "not_found",
  "message": "Error message" | null,
  "query": { /* the original query parameters */ }
}

Testing

The server comes with a comprehensive test suite using Vitest. Tests cover all available tools and include various scenarios including successful responses, empty results, and error handling.

Running Tests
npm test
Test Structure

The tests use Vitest's mocking capabilities to simulate Crossref API responses without making actual network requests. The test structure includes:

  1. Mock Data: Sample responses for title searches, author searches, and DOI lookups
  2. Mock Handlers: Testing versions of the handler functions in mcp-server-test-handlers.js
  3. Test Cases: Tests for all tools covering:
    • Successful API responses
    • Empty result sets
    • Error handling and network failures
Extending Tests

To add more test cases:

  1. Add new mock data to the test file if needed
  2. Create additional test cases in the relevant describe block
  3. Use the mockFetchResponse() helper to simulate API responses

Example:

it("should handle a new edge case", async () => {
  // Mock the response
  mockFetchResponse({
    // Your sample response data
  });

  // Call the handler
  const result = await handlers.searchByTitle({ title: "example" });

  // Assert the expected results
  expect(result).toMatchObject({
    // Expected response structure
  });
});

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.