DataCite MCP Server
Provides read-only access to DataCite's index of 125M+ research DOIs via natural language queries, enabling searching, metadata retrieval, citation formatting, and relationship exploration.
README
DataCite MCP Server
A Model Context Protocol (MCP) server that wraps the DataCite REST API, giving Claude and other MCP clients read-only access to DataCite's index of 125M+ research DOIs. No API key is required. The server runs as a local stdio process and communicates with Claude Desktop via the MCP protocol, enabling natural-language queries against the full DataCite catalogue: searching by keyword, person, funder, or repository; retrieving full metadata records; and exploring relationships between research outputs.
Claude Desktop Configuration
Add the following to your claude_desktop_config.json (found at ~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"datacite": {
"command": "node",
"args": ["/Users/[user_name]/Claude/projects/datacite/dist/index.js"]
}
}
}
Build Instructions
cd /Users/[user_name]/Claude/projects/datacite
npm install
npm run build
To inspect the server interactively:
npm run inspector
Available Tools
| Tool | Description |
|---|---|
search_dois |
Full-text search across DataCite's 125M+ DOI index with filters for resource type, funder, year, repository, and more. Supports cursor-based pagination. |
get_doi |
Retrieve complete metadata for a single DOI. Accepts any common DOI format (bare, URL, doi: prefix). Optionally includes the raw DataCite Schema XML. |
format_citation |
Format a DOI as a formatted citation string. Supports APA, MLA, Chicago, Harvard, IEEE, Vancouver, BibTeX, and CSL JSON. |
get_doi_metrics |
Retrieve view, download, and citation counts for a DOI. Optionally include time-series arrays. |
get_related_works |
Explore the relationship graph for a DOI — citations, references, versions, and parts. |
search_by_person |
Find all DOIs associated with a researcher by ORCID iD or name. |
list_repositories |
Browse DataCite member repositories (clients) with optional search and filtering. |
get_doi_schema_xml |
Fetch the raw DataCite Metadata Schema XML for a DOI (base64-decoded). |
Available Prompts
Prompts are pre-built workflows invocable from Claude Desktop's prompt menu (⌘K → "Use a prompt").
| Prompt | Arguments | Description |
|---|---|---|
find-top-works-by-topic |
resource_type (required), topic (required) |
Find the top 10 most relevant DataCite records for a resource type and subject area. Returns a ranked table with repository/year observations and a suggested follow-up. |
repository-summary |
repository_name (required) |
Full metadata quality and activity report for a repository. Pass a common name like "Zenodo" or "Dryad" — the prompt resolves it to a DataCite client_id automatically. |
researcher-profile |
identifier (required) |
Researcher profile from DataCite records. Pass an ORCID iD (0000-0001-8135-3489) for exact matching, or a name ("Jane Smith") for fuzzy search with disambiguation. |
Available Resources
| Resource URI | Description |
|---|---|
datacite://schema/resource-types |
The complete DataCite resourceTypeGeneral controlled vocabulary. |
datacite://providers |
All DataCite member providers (organisations). |
datacite://clients |
All DataCite repository clients. |
datacite://doi/{doi} |
Full metadata record for a given DOI as JSON. |
datacite://doi/{doi}/citations |
Works that cite the given DOI. |
datacite://doi/{doi}/references |
Works cited by the given DOI. |
datacite://provider/{id} |
Metadata for a specific DataCite provider. |
datacite://client/{id} |
Metadata for a specific DataCite repository client. |
Environment Variables
Copy .env.example to .env and adjust as needed. All variables are optional — defaults work out of the box.
| Variable | Default | Description |
|---|---|---|
MCP_USER_AGENT_URL |
https://github.com/alexwade/datacite-mcp |
URL included in the HTTP User-Agent header |
MCP_USER_AGENT_EMAIL |
(empty) | Contact email for DataCite's polite pool |
DATACITE_RATE_LIMIT_RPS |
10 |
Max API requests per second |
CACHE_DOI_TTL_SECONDS |
3600 |
DOI cache TTL (1 hour) |
CACHE_SEARCH_TTL_SECONDS |
300 |
Search cache TTL (5 minutes) |
CACHE_STATIC_TTL_SECONDS |
86400 |
Static data cache TTL (24 hours) |
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。