Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@@terranode-co/mcp-server What county is latitude 40.71, longitude -74.00 in?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for spatial queries via the Terranode Geospatial API. Gives AI agents (Claude Desktop, Cursor, etc.) tools to query geospatial datasets — point-in-polygon lookups, nearest feature search, distance calculations, and spatial joins.
Node.js 18+
A Terranode API key — sign up at app.terranode.co
Related MCP server: Overture Maps MCP Server
Add to your claude_desktop_config.json:
{
"mcpServers": {
"terranode": {
"command": "npx",
"args": ["@terranode-co/mcp-server"],
"env": {
"TERRANODE_API_KEY": "your-api-key-here"
}
}
}
}Add to your Cursor MCP settings (.cursor/mcp.json):
{
"mcpServers": {
"terranode": {
"command": "npx",
"args": ["@terranode-co/mcp-server"],
"env": {
"TERRANODE_API_KEY": "your-api-key-here"
}
}
}
}Add to your Claude Code MCP settings:
{
"mcpServers": {
"terranode": {
"command": "npx",
"args": ["@terranode-co/mcp-server"],
"env": {
"TERRANODE_API_KEY": "your-api-key-here"
}
}
}
}Tool | Description |
| List all available geospatial datasets (system + custom) |
| Point-in-polygon: which feature contains a coordinate? |
| Find the N nearest features to a coordinate, with distances |
| Geodesic distance between two points (meters + miles) |
| Enrich multiple coordinates with polygon attributes at once |
Try these in Claude Desktop or Cursor:
"What county is latitude 40.71, longitude -74.00 in?"
"What datasets are available?"
"Find the 3 nearest counties to Central Park"
"How close is my proposed site (40.7128, -74.006) to the nearest school (34.0522, -118.2437)?"
"Which states do these 5 warehouse locations fall in?" (paste coordinates)
"What ZIP code is 37.7749, -122.4194 in?"
The full API spec is available at docs.terranode.co/openapi.yaml (OpenAPI 3.1).
Report issues or feature requests: feedback@terranode.co
Calculate the exact geodesic distance between two points on Earth's surface. Uses the WGS84 ellipsoid model for high accuracy. No dataset needed. Returns distance in both meters and miles.
Inputs: lat1 (required), lng1 (required), lat2 (required), lng2 (required).
Check which feature in a polygon dataset contains a given coordinate (point-in-polygon query). Returns the properties of all matching features, or indicates no match. Requires latitude, longitude, and a dataset id (UUID) from list_datasets. Call list_datasets first to get the id. This is NOT a geocoder — it requires numeric coordinates, not addresses or place names.
Inputs: lat (required), lng (required), dataset (required).
Get location attributes (county, flood zone, school district, etc.) for a coordinate. Queries all available datasets in one call and returns what matched. No dataset id needed — this is the simplest way to find out what's at a location. Optional radius expands the search area (meters). This is NOT a geocoder — it requires numeric coordinates, not addresses or place names.
Inputs: lat (required), lng (required), radius.
Find the nearest counties, districts, ZIP codes, or other features to a coordinate. Use this for proximity queries, finding what's nearby, or ranking features by distance. Returns features sorted by distance (meters and miles), measured to feature boundary. Also works as a reverse lookup — returns the containing feature (distance=0) plus neighbors. Requires latitude, longitude, and a dataset id (UUID) from list_datasets.
Inputs: lat (required), lng (required), dataset (required), n, +1 more.
List all geospatial datasets available for querying. Returns pre-loaded public datasets (US states, counties, ZIP codes, etc.) and your custom uploaded datasets. Each dataset has an id (UUID) — always use the id when calling other tools.
Inputs: no named parameters.
Enrich a set of coordinates with attributes from a polygon dataset. For each point, returns the properties of the polygon it falls within. Like running check_location on multiple points at once, but more efficient. Requires an array of coordinate objects and a dataset id (UUID) from list_datasets. Call list_datasets first to get the id. Points need numeric coordinates, not addresses.
Inputs: points (required), dataset (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 6 tools
Tools have distinct purposes but check_location and enrich_location both perform point-in-polygon queries, which could cause confusion. However, clear descriptions differentiate their scopes (single vs. multi-dataset).
All tools follow a consistent verb_noun pattern with lowercase and underscores, e.g., calculate_distance, list_datasets, spatial_join.
Six tools cover essential geospatial operations without excess or deficiency, well-scoped for a geospatial analysis server.
Covers core geospatial queries (distance, point-in-polygon, enrichment, proximity) but lacks tools for dataset upload, detailed metadata, or buffer/area calculations. Minor gaps.
Geospatial AI MCP server — satellite imagery, embeddings, weather, GNS governance
MCP server for Japan geodata: cadastral lot numbers (chiban) and reverse geocoding, for AI agents.
Hosted MCP server for live public-data APIs and Skills for AI agents.
MCP server for querying Tulsa County, Oklahoma property and parcel data for AI agents.