Skip to content
Conduit

mcp server

openstreetmap-mcp

Advanced OpenStreetMap MCP Server for AI Agents. Goes beyond geocoding with Neighborhood Livability Scoring and Commute Analysis. Modular, stable, and Windows-ready.

neco001/openstreetmap-mcpcommunitylistedunreviewed
Use tool

Set up

Runs on your computer. Pick your agent and operating system.

Before you start

  • uv (`curl -LsSf https://astral.sh/uv/install.sh | sh`)
~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "openstreetmap-mcp": {
      "command": "uvx",
      "args": [
        "openstreetmap-mcp"
      ]
    }
  }
}

same format Cursor, GitHub Copilot CLI, Junie, JetBrains AI Assistant, Augment (Auggie CLI), Factory Droid, Kiro, Devin CLI +12

Values in angle brackets are placeholders. Restart Claude Desktop after editing its config.Claude Desktop docs

README

synced Oct 23, 2026

Enterprise OpenStreetMap (OSM) MCP Analytics Server

A modular, high-performance Model Context Protocol (MCP) server for OpenStreetMap. This implementation goes beyond simple geocoding, providing advanced geospatial analytics like neighborhood livability scoring and commute analysis.

🚀 Why this server?

While there are many OSM MCP servers, this version is built for stability, modular extension, and advanced insights:

  • Modular Architecture: Clean separation between API Client, Utility logic, and Tool definitions. Easy to audit and extend.
  • Enterprise Analytics:
    • analyze_neighborhood: Calculates a "Livability Score" based on walking proximity to essential services (groceries, healthcare, parks).
    • analyze_commute: Multi-modal comparison of travel times (car, bike, foot) for lifestyle planning.
  • Efficient: Uses FastMCP for asynchronous I/O and structured tool registration.
  • Windows Optimized: Built and tested to run reliably as a background process on Windows environments.

🛠 Features

Core Tools

  • Geocoding: geocode_address, reverse_geocode
  • Routing: get_route_directions (OSRM based)
  • Search: find_nearby_places, search_category
  • Analytics: explore_area, analyze_neighborhood, analyze_commute

Specialized Tools (Optional)

Found in tools/extras.py (not loaded by default for performance):

  • find_schools_nearby
  • find_ev_charging_stations
  • find_parking_facilities
  • suggest_meeting_point

Resources

  • location://place/{query}: Real-time place metadata.
  • location://map/{style}/{z}/{x}/{y}: Interactive map tile retrieval.

📦 Installation

Requirements

  • Python 3.10+
  • uv (recommended) or pip

Method 1: Via MCP Config (Claude/Cursor)

Add this to your mcp_config.json:

{
  "mcpServers": {
    "osm-mcp": {
      "command": "python",
      "args": [
        "c:/path/to/osm-mcp-server/src/openstreetmap_mcp/server.py"
      ],
      "env": {
        "PYTHONPATH": "c:/path/to/osm-mcp-server/src/openstreetmap_mcp"
      }
    }
  }
}

Method 2: Local Development

git clone https://github.com/neco001/openstreetmap-mcp
cd openstreetmap-mcp
uv sync

📂 Project Structure

src/openstreetmap_mcp/
├── server.py           # Main Entry Point
├── instance.py         # FastMCP lifecycle
├── client.py           # HTTP logic for OSM/OSRM/Overpass
├── utils.py            # Haversine & geometric helpers
├── tools/              # Categorized tool definitions
│   ├── geocoding.py
│   ├── routing.py
│   ├── search.py
│   └── analysis.py
└── resources.py        # Map & Data resources

⚖️ License

MIT License - feel free to use, modify and distribute.

Acknowledgments

Original logic & concepts by Jagan Shanmugam. This repository is a modular refactor focused on Enterprise usage, Windows compatibility, and Analytics tools.

↑↓ navigate openesc close