- Home
- MCP servers
- Geocontext
Geocontext
- typescript
4
GitHub Stars
typescript
Language
5 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
This MCP server provides spatial context services for use with language models, enriching answers with geographic data such as place names, coordinates, administrative information, cadastre details, and urban planning documents. It exposes spatial services that translate locations into coordinates, heights, and place-specific attributes, helping your assistant respond accurately to questions involving geography and maps.
How to use
You can connect to this MCP server in two ways: as a remote HTTP server or as a local STDIO server that runs inside your environment. The HTTP method lets your client reach the server through a standard URL. The STDIO method runs the MCP server as a local process and communicates via standard input/output. You can choose the method that best fits your deployment and latency needs.
With HTTP, point your MCP client at the server URL and send requests that invoke spatial services such as geocode, altitude, and information queries for a given location. The server processes these requests on the backend, leveraging the GeoPlatform services to return structured results like coordinates, elevations, and administrative attributes.
With STDIO, start the local server process and connect your MCP client to it through the standard streams. This mode is useful for local development or when you want tight coupling with your runtime environment. Ensure you pass the correct command and arguments to spawn the server process, then issue requests through the MCP protocol as you would with a remote server.
Security and transport preferences can be controlled through environment settings. A common option is TRANSPORT_TYPE, which selects between stdio and http transports. Use stdio for local runs and http when you want centralized, remote access to the MCP endpoints.
How to install
Prerequisites you need before installation: a modern Node.js environment with npm, and optionally Docker for containerized runs.
# Prerequisites
node -v
npm -v
# Optional: Docker is recommended for running in containers
docker -v
How to install
Option A: Run via STDIO (local development or embedding in another app) follow these steps.
# Step 1: Install dependencies
npm install
# Step 2: Build the project if a build step is required
npm run build
# Step 3: Start the STDIO MCP server (example command shown below)
npx -y @mborne/geocontext -- your-args-here
How to install
Option B: Run via HTTP (remote MCP server) follow these steps.
# Step 1: Install dependencies and start the HTTP server via Docker (if using Docker)
docker compose build
docker compose up -d
# Step 2: Use the HTTP MCP URL to connect from your MCP client
# Example endpoint is provided by the deployed service
Available tools
geocode
Converts a place name into coordinates (lon, lat) using the Geoplateforme autocompletion service.
altitude
Returns the altitude for a given (lon, lat) position using the Geoplateforme altitude service.
adminexpress
Fetches administrative information (municipality, department, region, etc.) for a location by its coordinates.
cadastre
Retrieves cadastral details (parcel, sheet, etc.) for a location by coordinates.
urbanisme
Fetches urban planning documents (PLU, POS, CC, PSMV) for a given location.
assiette_sup
Retrieves Public Utility Servitudes (SUP) information for a location.
gpf_wfs_list_types
Lists available WFS tables from the GeoPlatform WFS (GetCapabilities) (deprecated due to too many results).
gpf_wfs_search_types
Searches available WFS tables from the GeoPlatform WFS (GetCapabilities) by keywords.
gpf_wfs_describe_type
Retrieves the schema (DescribeFeatureType) of a WFS table.
gpf_wfs_get_features
Fetches features from a WFS table (GetFeature).