EEA GeoNetwork MCP Server
Enables interaction with the European Environment Agency (EEA) GeoNetwork Catalogue API to search, manage, and edit geographic metadata records. It supports advanced spatial searches, metadata exports in multiple formats, and authenticated updates to records and their attachments.
README
EEA GeoNetwork MCP Server
A Model Context Protocol (MCP) server that provides tools to interact with the European Environment Agency (EEA) GeoNetwork Catalogue API (GeoNetwork 4.4.9).
Features
This MCP server provides 20 tools for interacting with the EEA GeoNetwork Catalogue:
Search & Discovery
- search_records - Search for metadata records with full Elasticsearch query support
- search_by_extent - Find records by geographic bounding box
- get_record - Retrieve detailed metadata for a specific record by UUID
- get_record_by_id - Retrieve a record by its internal numeric ID
- get_related_records - Find related records (parent, children, services, datasets)
Data Export & Management
- get_record_formatters - List available export formats for a record
- export_record - Export metadata in various formats (XML, PDF, etc.)
- duplicate_record - Duplicate an existing metadata record (requires authentication)
Record Editing (Requires Authentication)
- update_record - Update any field using XPath (supports ISO 19139 and ISO 19115-3)
- update_record_title - Simplified tool to update a record's title (auto-detects schema)
- add_record_tags - Add tags/categories to a record
- delete_record_tags - Remove tags/categories from a record
Resource/Attachment Management
- upload_file_to_record - Upload a file from local filesystem directly to a metadata record (requires authentication)
- get_attachments - List all attachments/resources for a metadata record
- delete_attachment - Delete a specific attachment from a record (requires authentication)
Catalogue Information
- get_site_info - Get catalogue configuration and site information
- get_sources - List catalogue sources and sub-portals
- list_groups - List all user groups
- get_tags - Get available tags/categories
- get_regions - Get geographic regions/extents
Installation
Option 1: Docker (Recommended)
The easiest way to run the server is using Docker:
# Build and start the container
docker-compose up -d
# Check logs
docker-compose logs -f
# Stop the container
docker-compose down
The server will be available at http://localhost:3001
Environment Variables:
You can customize the configuration by creating a .env file or editing the docker-compose.yml:
environment:
- PORT=3001
- BASE_URL=https://galliwasp.eea.europa.eu/catalogue/srv/api
- CATALOGUE_USERNAME=your_username
- CATALOGUE_PASSWORD=your_password
Option 2: Manual Installation
- Install dependencies:
npm install
- Build the project:
npm run build
- (Optional) Configure environment variables:
# Create a .env file in the project root
PORT=3001
BASE_URL=https://galliwasp.eea.europa.eu/catalogue/srv/api
MAX_SEARCH_RESULTS=20
# Authentication for write operations (duplicate, update)
CATALOGUE_USERNAME=your_username
CATALOGUE_PASSWORD='your_password'
# Rate limiting
RATE_LIMIT_WINDOW_MS=900000
RATE_LIMIT_MAX_REQUESTS=100
Note: For passwords containing special characters ($, #, etc.), wrap the value in single quotes.
Usage
Streamable HTTP Server (MCP Transport)
The server uses the official MCP Streamable HTTP transport with Server-Sent Events (SSE).
Start the server:
npm start
The server will start on port 3001 (or the port specified in the PORT environment variable).
Available endpoints:
GET http://localhost:3001/health- Health check endpointGET http://localhost:3001/info- Server informationPOST http://localhost:3001/upload- Upload file to basket (multipart/form-data)GET http://localhost:3001/uploads/:filename- Retrieve uploaded filePOST http://localhost:3001/- MCP message endpoint (standard JSON-RPC)GET http://localhost:3001/- MCP SSE stream endpoint (for server-initiated messages)
Testing the server:
# Check if server is running
curl http://localhost:3001/health
# Should return: {"status":"ok","service":"eea-geonetwork-mcp"}
# Upload a file to the basket
curl -X POST http://localhost:3001/upload -F "file=@myfile.pdf"
# Returns: {"success":true,"file":{"url":"http://localhost:3001/uploads/myfile-123456789.pdf",...}}
Upload Basket
The server includes a built-in upload basket for temporary file storage. This allows you to upload files first, then attach them to metadata records using the URL.
Swagger UI (Easiest Method):
- Open
http://localhost:3001/api-docsin your browser - Expand the "POST /upload" endpoint
- Click "Try it out"
- Select a file and click "Execute"
- Copy the returned URL from the response
- Use this URL with the
upload_resource_from_urlMCP tool
How it works:
- Upload a file via
POST /uploadendpoint (or use Swagger UI) - Server stores the file in the
uploads/directory - Server returns a URL:
http://localhost:3001/uploads/filename - Use this URL with
upload_resource_from_urltool to attach to metadata records
Configuration:
# Optional environment variables
UPLOAD_DIR=./uploads # Upload directory (default: ./uploads)
MAX_FILE_SIZE=104857600 # Max file size in bytes (default: 100MB)
Swagger UI:
- Access the interactive API documentation and file upload interface at:
http://localhost:3001/api-docs
With Claude Desktop
Add to your Claude Desktop configuration file:
MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"eea-geonetwork": {
"command": "node",
"args": ["C:\\Users\\dubos\\_Projects\\EEA_sdi_mcp\\dist\\index.js"]
}
}
}
Example Queries
Once connected to Claude Desktop, you can ask questions like:
- "Search the EEA catalogue for datasets about air quality"
- "Find all metadata records within the bounding box of Europe"
- "Get detailed information about record UUID abc-123-def"
- "Export this metadata record as XML"
- "What geographic regions are available in the catalogue?"
- "Show me all datasets related to this parent record"
API Base URL
The server connects to: https://galliwasp.eea.europa.eu/catalogue/srv/api (Sandbox Environment)
Development
Commands
- Build:
npm run build- Compile TypeScript to dist/ - Watch mode:
npm run dev- Compile TypeScript in watch mode (auto-recompile on changes) - Start:
npm start- Run the compiled server
Development Workflow
For active development, run two commands in separate terminals:
Terminal 1 - Compile TypeScript in watch mode:
npm run dev
Terminal 2 - Start the server:
npm start
This way, TypeScript will automatically recompile when you make changes, and you can restart the server to pick up the new changes.
Architecture
The server uses the official MCP SDK with Streamable HTTP transport (stateless mode):
- Express server handles HTTP endpoints with CORS support
- MCP Server handles tool requests via Streamable HTTP/SSE
- Axios client communicates with the EEA GeoNetwork API (30s timeout)
- Modular design with separate files:
src/index.ts- Server setup and routingsrc/tools.ts- Tool definitions (20 tools)src/handlers.ts- Tool implementation handlerssrc/types.ts- TypeScript interfaces
Tools Reference
search_records
Search for metadata records in the catalogue.
Parameters:
query(string): Search textfrom(number): Starting position (default: 0)size(number): Number of results (default: 10, max: 100)bucket(string): Filter by facet bucketsortBy(string): Sort fieldsortOrder(string): "asc" or "desc"
get_record
Get detailed metadata for a specific record.
Parameters:
uuid(string, required): Record UUID or IDapproved(boolean): Only approved versions (default: true)
search_by_extent
Search records by geographic extent.
Parameters:
minx(number, required): Minimum longitude (west)miny(number, required): Minimum latitude (south)maxx(number, required): Maximum longitude (east)maxy(number, required): Maximum latitude (north)relation(string): Spatial relationship - "intersects", "within", or "contains" (default: "intersects")
export_record
Export a metadata record in a specific format.
Parameters:
uuid(string, required): Record UUIDformatter(string, required): Format identifier (e.g., "xml", "pdf", "full_view")
Use get_record_formatters first to see available formats for a record.
duplicate_record
Duplicate an existing metadata record with a new UUID. Requires authentication.
Parameters:
metadataUuid(string, required): UUID of the record to duplicategroup(string, optional): Target group for the duplicated recordisChildOfSource(boolean, optional): Set the source record as parent of the new record (default: false)targetUuid(string, optional): Specific UUID to use for the duplicated record (auto-generated if not provided)hasCategoryOfSource(boolean, optional): Copy categories from source record (default: true)
get_record_by_id
Retrieve a metadata record by its internal numeric ID. Useful for getting the UUID after a duplicate operation.
Parameters:
id(number, required): Internal numeric ID of the record
update_record
Update a metadata record field using XPath. Requires authentication.
Parameters:
uuid(string, required): UUID of the record to updatexpath(string, required): XPath to the element to updatevalue(string, required): New value (text for simple replacement, or full XML element)operation(string, optional): Operation type - "replace" (default), "add", or "delete"updateDateStamp(boolean, optional): Update the record's timestamp (default: true)
Common XPaths:
- ISO 19139 title:
gmd:identificationInfo/*/gmd:citation/gmd:CI_Citation/gmd:title/gco:CharacterString - ISO 19115-3 title:
mdb:identificationInfo/*/mri:citation/cit:CI_Citation/cit:title/gco:CharacterString
update_record_title
Simplified tool to update a record's title. Automatically detects the schema (ISO 19139 or ISO 19115-3) and uses the correct XPath. Requires authentication.
Parameters:
uuid(string, required): UUID of the record to updatetitle(string, required): New title for the record
add_record_tags
Add tags (categories) to a metadata record. Requires authentication.
Parameters:
uuid(string, required): UUID of the recordtags(array of numbers, required): Array of tag IDs to add
Use get_tags first to find available tag IDs.
delete_record_tags
Remove tags (categories) from a metadata record. Requires authentication.
Parameters:
uuid(string, required): UUID of the recordtags(array of numbers, required): Array of tag IDs to remove
upload_file_to_record
Upload a file from your local filesystem directly to a metadata record as an attachment. The file is uploaded as binary data to GeoNetwork. Requires authentication.
Parameters:
metadataUuid(string, required): UUID of the metadata record to attach the file tofilePath(string, required): Absolute path to the local file (e.g.,C:\Users\name\document.pdfor/home/user/document.pdf)visibility(string, optional): The sharing policy - "PUBLIC" or "PRIVATE" (default: PUBLIC)approved(boolean, optional): Use approved version or not (default: false)
Example:
metadataUuid: "43d7c186-2187-4bcd-8843-41e575a5ef56"
filePath: "C:\\Documents\\myfile.pdf"
Note: This tool directly uploads the file to GeoNetwork. The file must be accessible on the machine running the MCP server.
get_attachments
List all attachments/resources for a metadata record.
Parameters:
metadataUuid(string, required): UUID of the metadata recordsort(string, optional): Sort results by "type" or "name" (default: name)approved(boolean, optional): Use approved version or not (default: true)filter(string, optional): Filter pattern for attachment names (default: *)
Returns: Array of attachment objects with details like filename, type, URL, size, etc.
delete_attachment
Delete a specific attachment from a metadata record. Requires authentication.
Parameters:
metadataUuid(string, required): UUID of the metadata recordresourceId(string, required): The ID/filename of the resource to deleteapproved(boolean, optional): Use approved version or not (default: false)
Note: Use get_attachments first to find the exact resourceId of the attachment you want to delete.
License
MIT
Related Links
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.