Street View

Street View MCP is a local server that lets AI models fetch Street View imagery, save images, and build virtual tours. It exposes simple tools to obtain images by location, coordinates, or panorama ID, and to assemble those images into HTML tours for interactive exploration.

How to use

You run the Street View MCP locally and connect to it with your MCP client. Start by launching the MCP server, then call its tools to fetch images, open saved images, and create HTML tours.

To run the server via the recommended local workflow, use the runtime that your setup supports (a helper command is shown in the configuration). The standard local startup is a Python module invocation that starts listening on a host and port you choose.

Once the server is running, you can use the provided tools to fetch images by address, coordinates, or panorama ID, save them to your output directory, and generate HTML pages that present a sequence of images as a virtual tour. You can also open saved images in your system’s default image viewer.

How to install

Prerequisites you need before installation are Python 3.9 or newer and a working installation of the UV toolchain for running MCP servers.

1. Clone the Street View MCP repository and enter the directory.

git clone https://github.com/vlad-ds/street-view-mcp.git
cd street-view-mcp

2. Create and activate a virtual environment, then install dependencies in development mode.

uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
uv pip install -e ".[dev]"

3. Set up a Google Maps API key with Street View API enabled (you’ll use this as an environment variable named API_KEY). You may store it temporarily in your shell or place it in a .env file.

export API_KEY=your_api_key_here

Starting the MCP server and basic usage

Start the MCP server so it is reachable by your clients. Use the following command to launch the server locally on a chosen host and port.

python -m street_view_mcp.main --host 127.0.0.1 --port 8000

Using the built-in MCP tools

The server provides a set of tools you can call from your MCP client to fetch Street View imagery, retrieve metadata, open images locally, and assemble HTML tours.

Available tools

get_street_view

Fetches a Street View image by address, coordinates, or panorama ID and saves it to a file with specified size, heading, and pitch.

get_metadata

Retrieves metadata for a Street View panorama, including location and coordinates.

open_image_locally

Opens a saved Street View image in your system's default image viewer.

create_html_page

Creates an HTML page that displays multiple Street View images as a virtual tour with optional titles and descriptions.