Skip to content
↑↓navigate↵openescclose
Claude+372Whisper+228LangChain+168Codex+223NotebookLM+276DALL-E 3+192DeepL+249n8n+208Topaz Video AI+153LlamaIndex+161
OpenStreetMap MCP Server logo
#140 · of 445 · by weekly downloadsMCP server

OpenStreetMap MCP Server

Keyless OpenStreetMap access with composite tools for meeting points, commutes and neighbourhoods.

0
SaveVisit website ↗
Rank
#140
of 445 · by weekly downloads
Weekly downloads
1.2K
npm + PyPI, last 7 days
GitHub stars
221
Repository on GitHub

Add it to your client

Community
Install
uvx osm-mcp-server
mcpServers
{
  "mcpServers": {
    "openstreetmap-mcp": {
      "command": "uvx",
      "args": [
        "osm-mcp-server"
      ]
    }
  }
}
Transport
stdio
Runtime
Python
Auth
Not required
Tools
12
License
MIT
Stars
221
Tools
geocode_addressreverse_geocodefind_nearby_placesget_route_directionssearch_categorysuggest_meeting_pointexplore_areafind_schools_nearbyanalyze_commutefind_ev_charging_stationsanalyze_neighborhoodfind_parking_facilities

The interesting part is not the geocoder but the shape of the tool list. Alongside the primitives — geocode, reverse geocode, nearby places, directions, category search — sit tools that answer a whole question at once: suggest_meeting_point takes several people's locations and proposes where to meet, analyze_commute compares transport options between a home and a work address, and analyze_neighborhood scores an area for liveability. Chaining raw calls to reach those answers is exactly where models drift, so folding the chain into one tool removes a class of mistakes. There are also narrow lookups for schools, EV charging points and parking, plus map tiles and place data exposed as MCP resources. No API key is involved.

Comments(0)

Sign in to comment

No comments yet — be the first.

Similar MCP servers

Report this comment

Why are you reporting this?