Valhalla MCP Server
Integrates Claude Desktop with the Valhalla routing engine to provide OpenStreetMap-based routing and isochrone capabilities. It enables users to calculate travel paths and reachability polygons for various transportation modes using natural language.
README
Valhalla MCP Server
A Model Context Protocol (MCP) server that provides seamless integration between Claude Desktop and the Valhalla Open-Street-Map routing engine. Access professional routing and isochrone capabilities directly from your AI conversations without complex APIs.
Example Usage

Claude Desktop showing a route calculation between two points with detailed GeoJSON response from Valhalla MCP Server
Features
Currently Supported Valhalla Services
| MCP Tool/Resource | Valhalla Service | Description |
|---|---|---|
route tool |
/route |
Single origin → destination routing with alternatives |
isochrone tool |
/isochrone |
Travel time polygons showing reachable areas |
health resource |
/status |
Server status and version information |
tile resource |
/tile/{z}/{x}/{y} |
Vector tiles for client-side rendering |
Valhalla Services - Coming Soon
| Service | Description | Use Cases |
|---|---|---|
| Matrix | Distance/time matrices for multiple origins/destinations | Delivery optimization, logistics planning |
| Map-matching | Match GPS coordinates to road network | GPS trace cleaning, route correction |
| Elevation | Elevation profiles along routes or at points | Hiking routes, difficulty analysis |
| Expansion | Graph traversal visualization | Network analysis, reachability studies |
| Locate | Detailed metadata about nodes and edges | Address geocoding, road attributes |
| Centroid | Optimal convergence point from multiple locations | Meeting point optimization |
| Optimized Route | Multi-stop delivery routing with constraints | Logistics, delivery services |
Transportation Modes
auto- Car routingbicycle- Bicycle routingpedestrian- Walking routestaxi- Taxi routingbus- Public transit routing
Quick Start
Prerequisites
- Node.js 18+
- Claude Desktop
- NPM or Yarn package manager
Option 1: Local Valhalla Server
For production use or custom datasets:
# Prerequisites: Docker and Docker Compose
./start-mcp.sh
This script will:
- Build the MCP server
- Start local Valhalla with Monaco OSM data
- Wait for services to be ready
- Run integration tests
- Provide Claude Desktop configuration
Option 2: Use Your Pre-existing Valhalla Server
If you already have a Valhalla server running elsewhere:
# Clone and setup
git clone <repository-url>
cd valhalla-mcp
npm install
npm run build
# Configure environment variables
cp env.example .env
# Edit .env file and set VALHALLA_BASE_URL=http://your-valhalla-server:8002
Claude Desktop Integration
- Open Claude Desktop settings
- Add to your MCP configuration:
For Demo API (recommended for testing):
{
"mcpServers": {
"valhalla-mcp": {
"command": "node",
"args": ["dist/index.js"],
"cwd": "/YOUR/PATH/TO/valhalla-mcp",
"env": {
"VALHALLA_BASE_URL": "https://valhalla1.openstreetmap.de"
}
}
}
}
For Local Valhalla Server:
{
"mcpServers": {
"valhalla-mcp": {
"command": "node",
"args": ["dist/index.js"],
"cwd": "/YOUR/PATH/TO/valhalla-mcp",
"env": {
"VALHALLA_BASE_URL": "http://localhost:8002"
}
}
}
}
-
Update the
cwdpath to your actual project location -
Restart Claude Desktop
Test Your Installation
Try these commands in Claude Desktop to verify everything works:
Basic routing:
- "Calculate route from Monaco-Ville to Monte Carlo"
- "Get driving directions from 43.7384,7.4246 to 43.7396,7.4263"
- "Calculate bicycle route from 43.7350,7.4200 to 43.7450,7.4300"
Travel time analysis:
- "Show 10-minute drive isochrone from Monaco center"
- "Generate 15-minute bicycle travel polygon from 43.7311,7.4197"
You should receive detailed GeoJSON responses with route geometries and travel statistics!
Development Mode
For development with auto-reload:
npm run dev
Configuration
Environment Variables
The server uses environment variables for configuration. Create a .env file from the template:
cp env.example .env
Available environment variables:
VALHALLA_BASE_URL- Base URL for Valhalla service (default: http://localhost:8002)DEBUG- Enable debug logging (default: false)LOG_LEVEL- Log level: error, warn, info, debug (default: info)MCP_SERVER_NAME- MCP server name (default: valhalla-mcp-server)MCP_SERVER_VERSION- MCP server version (default: 0.1.0)
Common Valhalla endpoints:
http://localhost:8002- Local Docker instance (recommended)https://valhalla1.openstreetmap.de- Public demo API (✅ tested and working)https://your-server.com:8002- Custom deployment
Note: The main demo URL
https://valhalla.openstreetmap.deserves a web interface. Usehttps://valhalla1.openstreetmap.defor API access.
Docker Deployment
The complete stack can be deployed using Docker Compose:
# Build and start all services
docker-compose up -d
# View logs
docker-compose logs -f valhalla-mcp
# Stop services
docker-compose down
Usage Examples
Route Calculation
Request a route between two points:
{
"tool": "route",
"arguments": {
"origin": { "lat": 52.5200, "lon": 13.4050 },
"destination": { "lat": 52.5170, "lon": 13.3888 },
"mode": "bicycle",
"alternatives": 2,
"units": "kilometers"
}
}
Response includes GeoJSON LineString with route geometry and summary statistics:
{
"type": "FeatureCollection",
"features": [{
"type": "Feature",
"geometry": {
"type": "LineString",
"coordinates": [[13.4050, 52.5200], [13.3888, 52.5170]]
},
"properties": {
"distance_km": 2.1,
"duration_seconds": 420,
"duration_minutes": 7,
"mode": "bicycle"
}
}]
}
Isochrone Generation
Generate a 15-minute travel time polygon:
{
"tool": "isochrone",
"arguments": {
"origin": { "lat": 52.5200, "lon": 13.4050 },
"minutes": 15,
"mode": "pedestrian"
}
}
Health Check
Access server health information:
{
"resource": "health://status"
}
Integration with MCP Clients
Claude Desktop
Add to your Claude Desktop configuration:
{
"mcpServers": {
"valhalla": {
"command": "node",
"args": ["/path/to/valhalla-mcp/dist/index.js"],
"env": {
"VALHALLA_BASE_URL": "https://valhalla1.openstreetmap.de"
}
}
}
}
Note: Environment variables in Claude Desktop config override
.envfile values.
Other MCP Clients
The server implements the standard MCP protocol and works with any compliant client. Use stdio transport for local integration.
Architecture
┌─────────────────┐ MCP Protocol ┌─────────────────┐
│ MCP Client │ ◄─────────────────► │ Valhalla MCP │
│ (Claude, etc.) │ (stdio/HTTP) │ Server │
└─────────────────┘ └─────────┬───────┘
│ HTTP REST
┌─────────▼───────┐
│ Valhalla │
│ Routing Engine │
└─────────────────┘
Development
Installation
# Install dependencies
npm install
# Build the project
npm run build
Key dependencies:
@modelcontextprotocol/sdk- Official MCP SDKaxios- HTTP client for Valhalla API callszod- Runtime type validationgeojson- GeoJSON type definitionsdotenv- Environment variable management
Testing
# Run tests
npm test
# Run tests in watch mode
npm run test:watch
# Run linting
npm run lint
# Fix linting issues
npm run lint:fix
Type Checking
The project uses strict TypeScript configuration:
# Type check
npx tsc --noEmit
Performance
- Route calculation: < 200ms (local Valhalla instance)
- Isochrone generation: < 300ms
- Tile serving: < 30ms
- Health check: < 5ms
Performance depends on Valhalla configuration and available OSM data.
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests
- Submit a pull request
License
MIT License - see LICENSE file for details.
Acknowledgments
For issues and questions:
- Open an issue on GitHub
- Check the MCP documentation
- Review Valhalla documentation
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。