About osm-ui-puca
osm-ui-puca is a Streamlit web interface for the Púca MCP server, combining chat with OpenStreetMap-related tools. It is intended for people who want to submit queries through a UI or use the server with MCP clients and Pydantic AI.
Key features
- 01Streamlit chat interface
- 02OpenStreetMap-related tools
- 03Docker Compose setup
- 04Pydantic AI example
Use cases
- 01Querying OpenStreetMap-related information in a web UI
- 02Using Púca tools with Pydantic AI
MCP support
The Púca MCP server exposes tools for geocoding, distances, local amenities, Irish street names, vacant buildings, and Overpass results.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEosm-ui-puca
This repo has the python streamlit UI code for the Púca web interface. The web interface uses OpenAI and the Púca MCP server to provide tools related to OpenStreetMap.
Usage
This MCP server can be integrated with your client that supports MCP.

Docker
The docker-compose.yaml file included in this repo will run both the púca UI and MCP server so that you can enter queries into the UI.
To do so you will need to add your own OpenAI API Key in the .env file then start the containers with:
docker compose up -d
Check the logs for any issues and visit the web interface at http://your.ip.address:3301
Pydantic AI
See the example file client.py for how to use this MCP server with Python's pydantic_ai
Claude
Anthropic's Claude supports adding MCP servers if you edit the file:
vi ~/Library/Application\ Support/Claude/claude_desktop_config.json
adding the following but amending the URL:
{
"mcpServers": {
"puca": {
"command": "npx",
"args": [
"mcp-remote",
"http://192.168.1.40:3300/sse"
]
}
}
}
Tools
Below is a list of the tools made available via the MCP server:
- get_coordinates_for_address
- get_address_by_coordinates
- get_defibrillators
- get_distance_between_addresses
- get_distance_between_coords
- get_parking
- get_toilets
- get_post_offices
- get_cafes
- get_fast_food_places
- get_irish_street_names
- get_vacant_buildings
- get_overpass_results
Acknowledgements
This MCP server makes use of the freely available APIs from OSRM, Overpass and Nominatum. The data in OpenStreetMap is contributed by volunteers such as myself, so please don't forget to add attributions.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.