π
Google Maps MCP
Locationβ’β 640
Geocoding, place search, directions, and mapping services from Google.
Claude DesktopCursor
Overview
Geocoding, place search, directions, and mapping services from Google.
Setup
Run with npx:
npx -y @modelcontextprotocol/server-google-mapsConfiguration
GOOGLE_MAPS_API_KEY environment variableDocumentation
Google Maps MCP
Overview
Google Maps MCP is a Model Context Protocol server that enables AI agents to access Google Maps Platform services, including geocoding, place search, directions, and more. It allows agents to help with location-based queries, navigation, and geographic data.
Key Features
πΊοΈ Mapping Capabilities
- Geocoding: Convert addresses to coordinates and vice versa
- Place Search: Find places by name, type, or location
- Directions: Get routes between locations
- Distance Matrix: Calculate distances and travel times
- Place Details: Get detailed information about specific places
π Location Services
- Reverse Geocoding: Get address from coordinates
- Nearby Search: Find places near a location
- Autocomplete: Get place suggestions as you type
- Time Zones: Convert between time zones
π Navigation
- Route Planning: Get directions with multiple stops
- Travel Modes: Driving, walking, cycling, transit
- Traffic Awareness: Consider current traffic conditions
- Route Optimization: Find optimal routes for multiple destinations
Installation
# Using npx (recommended)
npx -y @modelcontextprotocol/server-google-maps
# With API key
GOOGLE_MAPS_API_KEY=your-api-key npx -y @modelcontextprotocol/server-google-maps
Configuration
Required Setup
-
Get a Google Maps API Key:
- Visit Google Cloud Console
- Create a new project or select existing
- Enable required APIs:
- Maps JavaScript API
- Places API
- Directions API
- Geocoding API
- Create API key with appropriate restrictions
-
Set Environment Variable:
export GOOGLE_MAPS_API_KEY=your-api-key-here -
Configure in Claude Desktop:
{ "mcpServers": { "google-maps": { "command": "npx", "args": ["-y", "@modelcontextprotocol/server-google-maps"], "env": { "GOOGLE_MAPS_API_KEY": "your-api-key-here" } } } }
Available Tools
| Tool | Description |
|---|---|
geocode | Convert address to coordinates |
reverse_geocode | Convert coordinates to address |
search_places | Search for places by query |
get_place_details | Get detailed info about a place |
get_directions | Get directions between locations |
distance_matrix | Calculate distances and times |
find_place | Find a place by text query |
places_nearby | Find places near a location |
timezone | Get timezone for coordinates |
Usage Examples
Geocode an Address
What are the coordinates for 1600 Amphitheatre Parkway, Mountain View, CA?
Returns:
Latitude: 37.4224764
Longitude: -122.0842499
Formatted Address: 1600 Amphitheatre Pkwy, Mountain View, CA 94043, USA
Reverse Geocode
What's at coordinates 40.7128Β° N, 74.0060Β° W?
Returns:
Address: New York, NY, USA
Place Type: locality
Search for Places
Find coffee shops near San Francisco
Returns:
1. Blue Bottle Coffee - 0.3 miles away
2. Sightglass Coffee - 0.5 miles away
3. Ritual Coffee Roasters - 0.7 miles away
...
Get Directions
Give me directions from San Francisco to Los Angeles
Returns:
Route: San Francisco, CA β Los Angeles, CA
Distance: 383 miles (616 km)
Duration: 5h 45m (via I-5 S)
Traffic: Moderate
Multi-Stop Route
Plan a route: Home β Coffee Shop β Office β Gym β Home
Returns:
Optimized Route:
1. Home (start)
2. Coffee Shop (5 min)
3. Office (15 min)
4. Gym (10 min)
5. Home (20 min)
Total: 50 minutes, 12 miles
Place Details
Tell me about the Eiffel Tower
Returns:
Name: Eiffel Tower
Address: Champ de Mars, 5 Avenue Anatole France, 75007 Paris, France
Rating: 4.6/5 (150,000+ reviews)
Type: Landmark, Tourist Attraction
Hours: Open 24 hours
Website: https://www.toureiffel.paris/
Distance Matrix
How long does it take to drive from these locations to downtown?
- Airport
- Train Station
- Hotel
Returns:
Airport β Downtown: 35 min (22 miles)
Train Station β Downtown: 15 min (8 miles)
Hotel β Downtown: 10 min (4 miles)
Integration with AI Agents
Travel Planning
Claude: "Let me help you plan your trip to Tokyo..."
[Searches for hotels, restaurants, attractions]
Claude: "Here's a suggested itinerary..."
[Provides directions between locations]
Local Business Research
User: "Find me a good Italian restaurant for a business dinner"
Claude: "Let me search for Italian restaurants in your area..."
[Searches places, gets details, checks ratings]
Claude: "I found 3 highly-rated options. Here are the details..."
Logistics Planning
User: "I need to deliver packages to 5 locations today"
Claude: "Let me optimize your delivery route..."
[Calculates optimal route using distance matrix]
Claude: "Here's the most efficient order to visit all locations..."
Real Estate Assistance
User: "What's the commute from this neighborhood to downtown?"
Claude: "Let me check the commute times..."
[Geocodes locations, gets directions]
Claude: "Average commute is 35 minutes by car, 50 minutes by transit..."
Custom Agent Integration
from mcp import ClientSession, StdioServerParameters
from mcp.client.stdio import stdio_client
import os
# Connect to Google Maps MCP
server_params = StdioServerParameters(
command="npx",
args=["-y", "@modelcontextprotocol/server-google-maps"],
env={"GOOGLE_MAPS_API_KEY": os.environ["GOOGLE_MAPS_API_KEY"]}
)
async with stdio_client(server_params) as (read, write):
async with ClientSession(read, write) as session:
await session.initialize()
# Geocode
result = await session.call_tool(
"geocode",
arguments={"address": "1600 Amphitheatre Parkway, Mountain View, CA"}
)
# Search places
places = await session.call_tool(
"search_places",
arguments={"query": "coffee shops near San Francisco"}
)
# Get directions
directions = await session.call_tool(
"get_directions",
arguments={
"origin": "San Francisco, CA",
"destination": "Los Angeles, CA",
"mode": "driving"
}
)
Google Maps API Pricing
| Service | Free Tier | Paid Tier |
|---|---|---|
| Geocoding | 40,000/month | $5 per 1,000 |
| Places | 128,000/month | $17 per 1,000 |
| Directions | 40,000/month | $5 per 1,000 |
| Distance Matrix | 100,000/month | $5 per 1,000 |
Google provides a $200 monthly credit automatically.
Pros
- β Comprehensive Data: Access to Google's massive place database
- β Accurate Geocoding: Industry-leading address-to-coordinate conversion
- β Rich Place Details: Reviews, hours, photos, and more
- β Multiple Travel Modes: Driving, walking, cycling, transit
- β Real-Time Traffic: Current traffic conditions
- β Global Coverage: Works worldwide
Cons
- β API Key Required: Must set up Google Cloud project
- β Cost at Scale: Can get expensive for high usage
- β Quota Limits: Daily and monthly request limits
- β Complex Setup: Multiple APIs to enable
- β Billing Required: Must have billing account for production use
When to Use
Choose Google Maps MCP when:
- You need accurate geocoding and mapping
- You want access to place information and reviews
- You need routing and directions
- You're building location-based applications
Consider alternatives when:
- You need free unlimited access (try OpenStreetMap-based solutions)
- You only need basic geocoding (simpler APIs available)
- You're on a tight budget (free tiers may be insufficient)
Resources
- GitHub: https://github.com/modelcontextprotocol/servers/tree/main/src/google-maps
- Google Maps Platform: https://mapsplatform.google.com/
- API Documentation: https://developers.google.com/maps
- Pricing: https://mapsplatform.google.com/pricing/
Comparison
| Feature | Google Maps MCP | OpenStreetMap | Mapbox MCP |
|---|---|---|---|
| Data Quality | β Excellent | β οΈ Variable | β Excellent |
| Coverage | β Global | β Global | β Global |
| Free Tier | β οΈ Limited | β Unlimited | β οΈ Limited |
| Place Details | β Rich | β οΈ Basic | β Good |
| Setup Complexity | β οΈ Moderate | β Simple | β οΈ Moderate |
| Cost at Scale | β οΈ High | β Free | β οΈ Moderate |
Last updated: May 2026
