google-maps

This MCP server exposes Google Maps Platform APIs (Places, Directions, Geocoding, Distance Matrix, Roads, Elevation, and route safety factors) through a standard MCP interface. It provides tools like search_places, get_place_details, get_directions, get_traffic_conditions, geocode_address, reverse_geocode, calculate_distance_matrix, snap_to_roads, get_speed_limits, get_route_elevation_gain, and calculate_route_safety_factors. Use the MCP inspector or your MCP client to discover and call these tools with the server's endpoint, passing your Google Maps API key via environment variable GOOGLE_MAPS_API_KEY. The server is designed for robust production use with structured logging, retries, and thorough input validation via Pydantic, and it can be deployed locally with uv or in containers.

To call a tool, connect your MCP client to the server and issue a tool invocation such as asking for nearby places, a route with traffic details, or geocoding coordinates. For Claude Desktop or other MCP clients, you can configure the server under mcpServers.google-maps with the appropriate command (uvx or google-maps-mcp-server) and environment variable, then query the included tools to retrieve and process location data in your AI workflows.

Prerequisites:

• Python 3.10 or higher (3.14+ recommended)
• uv package manager
• Google Maps API key with access to required APIs (Places, Directions, Geocoding, Distance Matrix, Roads)

Install using uv (recommended):

uv pip install google-maps-mcp-server

Install with pip (alternative):

pip install google-maps-mcp-server

Install from source (optional):

git clone https://github.com/ettysekhon/google-maps-mcp-server.git
cd google-maps-mcp-server
uv sync

Run locally to start the server (example):

# Using uvx as shown in MCP usage
uvx google-maps-mcp-server

Configure environment (example):

export GOOGLE_MAPS_API_KEY=your_maps_api_key_here

Optionally, set a .env file in your working directory with:

GOOGLE_MAPS_API_KEY=your_maps_api_key_here

Tips and notes:

• Ensure your GOOGLE_MAPS_API_KEY has the necessary API enablements (Places, Directions, Geocoding, Distance Matrix, Roads).
• If deploying in production, consider using a containerized deployment and a managed key secret store for GOOGLE_MAPS_API_KEY.
• The MCP server supports structured logging and exponential backoff retries; adjust LOG_LEVEL in your environment for more or less verbosity.
• When using Claude or other MCP clients, the server can be added with command uvx or google-maps-mcp-server; remember to provide the GOOGLE_MAPS_API_KEY in env so clients can authenticate API calls.
• The server exposes 11 tools across the Maps APIs; see the Quick Start section for real-world usage examples and how to verify the inspector shows all tools.