Model Context Protocol
MobilityTwin.Brussels exposes an MCP server that lets AI assistants — like Claude or ChatGPT — directly discover, authenticate, and query Brussels' mobility datasets through a standardized protocol.
No manual setup needed
Your AI assistant handles everything autonomously. Just ask it to query Brussels mobility data — it will create an account on your behalf, obtain an API token, discover available datasets, and fetch the data you need, all within the conversation. You never need to visit this website or write any code.
Available Tools
list_datasets
List all available dataset groups and their endpoints — names, formats, and descriptions.
No parameters required
get_dataset_info
Get detailed metadata for a specific endpoint — query parameters, format, and API path.
register
The AI creates an account on behalf of the user and receives an API token — no need to visit the website.
get_token
The AI authenticates with existing credentials and retrieves an API token to use in subsequent queries.
query_data
Query a dataset — single-point by timestamp or time range for supported endpoints.
get_latest_data
Fetch the most recent data from any endpoint — no timestamp needed.
Connect
Claude Code
Add to .mcp.json in your project root:
{
"mcpServers": {
"mobilitytwin": {
"type": "http",
"url": "https://mobilitytwin.brussels/mcp"
}
}
}
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"mobilitytwin": {
"type": "http",
"url": "https://mobilitytwin.brussels/mcp"
}
}
}
Any MCP Client
Point any MCP-compatible client to the endpoint:
https://mobilitytwin.brussels/mcp
Supports Streamable HTTP transport (MCP spec).
Example Workflow
Discover available data
The AI agent calls list_datasets to see all groups and endpoints, then get_dataset_info for details on a specific one.
Authenticate on your behalf
The agent creates an account for you via register or logs in via get_token — you never leave the chat.
Query data
Using the token, the agent calls query_data with a timestamp or time range, or get_latest_data for the most recent snapshot.
Analyze & respond
The agent processes the returned JSON — GeoJSON, GTFS, or raw data — and provides insights, visualizations, or answers directly in the conversation.