OSM Edit MCP Server
Enables AI assistants to search, validate, and edit OpenStreetMap data through natural language commands and built-in safety protections. It supports discovery of nearby amenities, geographic data exploration, and secure map editing via OAuth authentication.
README
OSM Edit MCP Server
A powerful Model Context Protocol (MCP) server that enables AI assistants to interact with OpenStreetMap data. Read, search, validate, and edit map data safely with built-in protections.
π What Can You Do?
- π Search Places: Find restaurants, cafes, hospitals, schools, and more
- π Validate Locations: Check coordinates and get detailed location info
- πΊοΈ Explore Areas: Discover what's in any geographic region
- βοΈ Edit Safely: Make map edits on the development server first
- π€ Natural Language: Use plain English to describe what you want
π¦ Prerequisites
- Python 3.10+
- (Optional) uv for fast dependency management
# Install uv (optional but recommended) curl -LsSf https://astral.sh/uv/install.sh | sh
π Quick Start (5 Minutes)
1οΈβ£ Install
git clone https://github.com/skywinder/osm-edit-mcp
cd osm-edit-mcp
uv sync --dev # Installs both base and development dependencies
2οΈβ£ Configure
cp .env.example .env
# No need to edit - defaults are ready to use!
3οΈβ£ Test
uv run python status_check.py
4οΈβ£ Connect to MCP Client
Important: MCP servers communicate via stdin/stdout with MCP clients. Don't run main.py directly!
Instead, configure the server in your MCP client:
- Cursor IDE: Settings β Features β MCP
- Claude Desktop: See MCP Client Setup
- VSCode (Cline): Add to settings.json
To test functionality without a client:
uv run python test_comprehensive.py
π Enable Write Operations (Optional)
To create or edit map data, you need OAuth authentication:
Step 1: Create Dev Account
Visit https://api06.dev.openstreetmap.org and sign up (separate from main OSM).
Step 2: Create OAuth App
- Go to your dev account settings β OAuth 2 Applications
- Register new application:
- Name:
OSM Edit MCP Dev - Redirect URI:
https://localhost:8080/callback - Permissions: Select all checkboxes
- Name:
Step 3: Add Credentials
Edit .env and add your OAuth credentials:
OSM_DEV_CLIENT_ID=your_client_id_here
OSM_DEV_CLIENT_SECRET=your_client_secret_here
Step 4: Authenticate
uv run python oauth_auth.py
Step 5: Verify
uv run python test_comprehensive.py
Expected: β 19/19 tests passing
π Available Tools
π Search & Discovery
| Tool | Description | Example |
|---|---|---|
find_nearby_amenities |
Find places around a location | "Find restaurants within 500m" |
get_place_info |
Search places by name | "Where is Central Park?" |
search_osm_elements |
Text search for any element | "Search for coffee shops" |
smart_geocode |
Convert address to coordinates | "10 Downing Street, London" |
π Location Tools
| Tool | Description | Example |
|---|---|---|
validate_coordinates |
Check if coordinates are valid | 51.5074, -0.1278 |
get_osm_elements_in_area |
Get all elements in a box | "What's in this area?" |
get_osm_statistics |
Area statistics | "How many restaurants?" |
πΊοΈ OSM Data Access
| Tool | Description | Example |
|---|---|---|
get_osm_node |
Get node by ID | Node details |
get_osm_way |
Get way by ID | Street/building info |
get_osm_relation |
Get relation by ID | Complex features |
βοΈ Editing Tools (Requires Auth)
| Tool | Description | Example |
|---|---|---|
create_changeset |
Start editing session | Required for edits |
create_osm_node |
Add new point | "Add restaurant here" |
create_place_from_description |
Natural language creation | "Add coffee shop called Bean There at..." |
π‘ Usage Examples
Find Nearby Restaurants
# Find Italian restaurants near the Colosseum
result = await find_nearby_amenities(
lat=41.8902, lon=12.4922,
radius_meters=500,
amenity_type="restaurant"
)
Validate Coordinates
# Check if coordinates are valid and get location info
result = await validate_coordinates(51.5074, -0.1278)
# Returns: "London, England, United Kingdom"
Natural Language Search
# Parse natural language requests
result = await parse_natural_language_osm_request(
"Find coffee shops near the Eiffel Tower"
)
π₯οΈ MCP Client Integration
Quick Setup for Popular Clients
<details> <summary><b>Cursor IDE</b></summary>
// With uv (Recommended)
{
"mcpServers": {
"osm-edit": {
"command": "uv",
"args": ["run", "python", "main.py"],
"cwd": "/path/to/osm-edit-mcp",
"env": {
"OSM_USE_DEV_API": "true",
"LOG_LEVEL": "INFO"
}
}
}
}
// Alternative: Using wrapper script (if uv has path issues)
{
"mcpServers": {
"osm-edit": {
"command": "/path/to/osm-edit-mcp/run_mcp.sh",
"args": [],
"env": {
"OSM_USE_DEV_API": "true",
"LOG_LEVEL": "INFO"
}
}
}
}
Add to Cursor Settings β Features β MCP </details>
<details> <summary><b>Claude Desktop</b></summary>
// With uv (Recommended)
{
"mcpServers": {
"osm-edit": {
"command": "uv",
"args": ["run", "python", "main.py"],
"cwd": "/path/to/osm-edit-mcp",
"env": {
"OSM_USE_DEV_API": "true",
"LOG_LEVEL": "INFO"
}
}
}
}
// Alternative: Using wrapper script (if uv has path issues)
{
"mcpServers": {
"osm-edit": {
"command": "/path/to/osm-edit-mcp/run_mcp.sh",
"args": [],
"env": {
"OSM_USE_DEV_API": "true",
"LOG_LEVEL": "INFO"
}
}
}
}
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (Mac)
</details>
<details> <summary><b>Continue.dev</b></summary>
{
"mcpServers": [
{
"name": "osm-edit",
"command": "uv",
"args": ["run", "python", "main.py"],
"cwd": "/path/to/osm-edit-mcp"
}
]
}
Add to ~/.continue/config.json
</details>
<details> <summary><b>Cline (VSCode)</b></summary>
{
"cline.mcpServers": {
"osm-edit": {
"command": "uv",
"args": ["run", "python", "main.py"],
"cwd": "./osm-edit-mcp"
}
}
}
Add to VSCode settings or .vscode/settings.json
</details>
π Full MCP Client Setup Guide - Detailed instructions for all clients
Example Queries
- "Find restaurants near Times Square"
- "What's at coordinates 48.8584, 2.2945?"
- "Search for hospitals in Seattle"
π‘οΈ Safety Features
- β Development API by default - Safe testing environment
- β OAuth protection - Edits require authentication
- β Rate limiting - Respects API limits
- β Input validation - Prevents invalid data
- β Changeset management - Groups edits properly
π Project Status
- Version: 0.1.0
- Tests: 100% passing (19/19)
- Python: 3.10+
- License: MIT
π Remote Deployment (Make it Accessible Anywhere)
The OSM Edit MCP Server can be deployed as a web service accessible from anywhere. This is useful for:
- Team collaboration
- Integration with web applications
- Running on cloud servers
- Access from multiple devices
π Quick Deploy with Docker
1. Prerequisites
- Docker and docker-compose installed
- A server with public IP or domain name
- SSL certificate (or use the self-signed cert for testing)
2. Deploy Steps
# Clone the repository
git clone https://github.com/skywinder/osm-edit-mcp
cd osm-edit-mcp
# Configure environment
cp .env.example .env
# Edit .env with your OAuth credentials and API_KEY
# Deploy with Docker
chmod +x deploy.sh
./deploy.sh
The deploy script will:
- Build Docker containers
- Generate SSL certificates (self-signed for development)
- Start the web server on port 8000
- Set up Nginx reverse proxy on port 443
3. Access Your Server
After deployment, access your server at:
https://your-server-ip/(with Nginx SSL)http://your-server-ip:8000/(direct access)- API docs:
http://your-server-ip:8000/docs
π‘ API Usage
All MCP functionality is exposed via REST API endpoints. Authenticate with your API key:
# Example: Find nearby amenities
curl -X POST https://your-server-ip/api/nearby-amenities \
-H "Authorization: Bearer your-api-key-here" \
-H "Content-Type: application/json" \
-d '{
"lat": 51.5074,
"lon": -0.1278,
"radius_meters": 500,
"amenity_type": "restaurant"
}'
π Security Configuration
- API Key: Set a strong
API_KEYin your.envfile - SSL Certificate: Replace self-signed cert with a real one for production
- Firewall: Only expose necessary ports (80, 443)
- Rate Limiting: Configured via
RATE_LIMIT_PER_MINUTEin.env
βοΈ Cloud Platform Deployment
<details> <summary><b>Deploy to AWS EC2</b></summary>
# Launch EC2 instance (Ubuntu 22.04 recommended)
# Install Docker
sudo apt update
sudo apt install docker.io docker-compose
# Clone and deploy
git clone https://github.com/skywinder/osm-edit-mcp
cd osm-edit-mcp
sudo ./deploy.sh
</details>
<details> <summary><b>Deploy to DigitalOcean</b></summary>
# Create a Droplet with Docker pre-installed
# SSH into your droplet
ssh root@your-droplet-ip
# Clone and deploy
git clone https://github.com/skywinder/osm-edit-mcp
cd osm-edit-mcp
./deploy.sh
</details>
<details> <summary><b>Deploy to Google Cloud Run</b></summary>
# Build and push to Container Registry
gcloud builds submit --tag gcr.io/PROJECT-ID/osm-edit-mcp
# Deploy to Cloud Run
gcloud run deploy osm-edit-mcp \
--image gcr.io/PROJECT-ID/osm-edit-mcp \
--platform managed \
--allow-unauthenticated \
--set-env-vars API_KEY=your-api-key
</details>
π§ Advanced Configuration
Custom Domain & SSL
# Update nginx.conf with your domain
server_name yourdomain.com;
# Use Let's Encrypt for free SSL
sudo apt install certbot python3-certbot-nginx
sudo certbot --nginx -d yourdomain.com
Environment Variables
All configuration is done via environment variables. Key settings:
OSM_USE_DEV_API: Use dev (true) or production (false) APIAPI_KEY: Authentication key for API accessRATE_LIMIT_PER_MINUTE: API rate limitingLOG_LEVEL: Logging verbosity
Monitoring
# View logs
docker-compose logs -f
# Check health
curl https://your-server/health
# Monitor resources
docker stats
π Production Checklist
- [ ] Use production OSM API (
OSM_USE_DEV_API=false) - [ ] Set strong
API_KEY - [ ] Install real SSL certificate
- [ ] Configure firewall rules
- [ ] Set up monitoring/alerts
- [ ] Enable automated backups
- [ ] Configure log rotation
- [ ] Set resource limits in docker-compose.yml
π§ͺ Testing
# Quick test
python quick_test.py
# Full test suite
python test_comprehensive.py
# Check your edits
# Visit: https://api06.dev.openstreetmap.org/user/YOUR_USERNAME/history
π¨ Troubleshooting
| Issue | Solution |
|---|---|
| "Server hangs" when running main.py | This is normal! MCP servers wait for client input. Use uv run python test_comprehensive.py instead |
| "401 Unauthorized" | Run uv run python oauth_auth.py |
| "Client auth failed" | Check OAuth credentials in .env |
| Import errors | Run uv sync --dev |
| Can't see changesets | Check dev server URL (not main OSM) |
| uv: command not found | Install uv: curl -LsSf https://astral.sh/uv/install.sh | sh |
| How do I use the server? | Configure in MCP client or run uv run python explain_mcp_server.py |
π Documentation
- Quick Reference Card - All commands on one page
- MCP Client Setup - Cursor, Claude, VSCode, etc.
- Running the Server - Background, monitoring, auto-restart
- Quick Start Guide
- API Examples
- OSM Tagging Guide
- Contributing
- Security Policy
π€ Contributing
We welcome contributions! See CONTRIBUTING.md for guidelines.
π Links
Ready to explore the world's map data? Start with the Quick Start above! π
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.