Street View MCP
Vista de calle MCP
Un servidor de protocolo modelo-cliente (MCP) para la API de Google Street View que permite que los modelos de IA obtengan y muestren imágenes de vistas de calles y creen recorridos virtuales.
Uso con Claude Desktop
Para utilizar Street View MCP con Claude Desktop:
Asegúrese de tener instalado
uv: Guía de instalación de UVClonar este repositorio:
git clone https://github.com/vlad-ds/street-view-mcp.git cd street-view-mcpInstalar dependencias:
uv pip install -e ".[dev]"Obtenga una clave API de Google Maps (instrucciones a continuación)
Agregue lo siguiente a su archivo
claude_desktop_config.jsonde Claude Desktop:
"street_view": {
"command": "uv",
"args": [
"run",
"--directory",
"/path/to/street-view-mcp", // Replace with your actual path
"mcp",
"run",
"src/street_view_mcp/server.py"
],
"env": {
"API_KEY": "your_google_maps_api_key_here" // Add your API key here
}
}Después de la configuración, puedes usar Street View MCP en Claude Desktop simplemente escribiendo "/street_view".
Related MCP server: MCP Google Map Server
Descripción general
Street View MCP proporciona una interfaz sencilla para que los modelos de IA puedan:
Obtener imágenes de Street View por dirección, coordenadas o ID de panorama
Guardar imágenes en archivos locales
Abrir imágenes guardadas en el visor predeterminado
Cree páginas HTML que compilen múltiples imágenes de Street View en recorridos virtuales
Requisitos
Python 3.9+
Clave API de Google Maps con API de Street View habilitada
paquete
fastmcpgestor de paquetes
uv(recomendado)
Instalación
# Clone the repository
git clone https://github.com/vlad-ds/street-view-mcp.git
cd street-view-mcp
# Create and activate a virtual environment with uv (recommended)
uv venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
# Install dependencies
uv pip install -e ".[dev]"Configuración de la clave API
El MCP de Street View requiere una clave API de Google Maps con la API de Street View habilitada:
Visita la consola de Google Cloud
Crea un nuevo proyecto o selecciona uno existente
Habilite la "API estática de Street View" en la biblioteca de API
Cree una clave API desde la página Credenciales
Establezca la clave API como una variable de entorno:
# Set temporarily in your shell:
export API_KEY=your_api_key_here
# Or create a .env file in the project root:
echo "API_KEY=your_api_key_here" > .envUso
Iniciando el servidor MCP
python -m street_view_mcp.main --host 127.0.0.1 --port 8000El servidor estará disponible para los modelos de IA en el host y puerto especificados.
Uso como herramienta CLI
# Fetch Street View image by address
python -m street_view_mcp.street_view --address "Empire State Building, NY" --output output/empire_state.jpg
# Fetch Street View image by latitude/longitude
python -m street_view_mcp.street_view --latlong "40.748817,-73.985428" --output output/coords.jpg --heading 180
# Fetch Street View image by panorama ID
python -m street_view_mcp.street_view --pano PANO_ID --output output/panorama.jpgHerramientas MCP
El MCP de Street View proporciona las siguientes herramientas para modelos de IA:
get_street_view
Obtiene una imagen de Street View según la ubicación, las coordenadas o el ID del panorama y la guarda en un archivo.
{
"filename": "empire_state.jpg",
"location": "Empire State Building, NY",
"size": "600x400",
"heading": 90,
"pitch": 10
}Parámetros:
filename(obligatorio): nombre para guardar la imagen (no debe existir ya)location(opcional): Dirección para obtener la imagenlat_lng(opcional): coordenadas separadas por comas (p. ej., "40.748817,-73.985428")pano_id(opcional): ID de panorama específicosize(opcional): Dimensiones de la imagen como "ancho x alto" (predeterminado: "600x400")heading(opcional): rumbo de la cámara en grados (0-360, predeterminado: 0)pitch(opcional): Inclinación de la cámara en grados (-90 a 90, predeterminado: 0)fov(opcional): Campo de visión en grados (10-120, predeterminado: 90)radius(opcional): radio de búsqueda en metros (predeterminado: 50)source(opcional): Fuente de la imagen ("predeterminada" o "exterior", predeterminada: "predeterminada")
Nota: Se debe proporcionar exactamente uno de los valores: location , lat_lng o pano_id .
get_metadata
Obtiene metadatos sobre un panorama de Street View.
{
"location": "Empire State Building, NY"
}Parámetros:
Los mismos parámetros de ubicación que
get_street_viewDevuelve metadatos JSON con estado, derechos de autor, fecha, ID de panorama y coordenadas
open_image_locally
Abre una imagen de Street View guardada en la aplicación predeterminada.
{
"filename": "empire_state.jpg"
}Parámetros:
filename(obligatorio): el nombre del archivo de la imagen a abrir (debe existir en el directorio de salida)
create_html_page
Crea una página HTML que muestra múltiples imágenes de Street View como un recorrido virtual.
{
"filename": "nyc_tour.html",
"title": "New York City Tour",
"html_elements": [
"<h1>New York City Landmarks Tour</h1>",
"<p>Explore famous landmarks through Street View images.</p>",
"<h2>Empire State Building</h2>",
"<img src='../output/empire.jpg' alt='Empire State Building'>",
"<p class='location'>350 Fifth Avenue, New York, NY</p>",
"<p class='description'>This 102-story Art Deco skyscraper was completed in 1931.</p>"
]
}Parámetros:
html_elements(obligatorio): Lista de elementos de contenido HTMLfilename(obligatorio): nombre del archivo HTMLtitle(opcional): Título de la página (predeterminado: "Recorrido por Street View")
Importante: Al referenciar imágenes, utilice siempre la ruta ../output/filename.jpg .
Creación de recorridos virtuales
El MCP de Street View permite la creación de recorridos virtuales combinando múltiples imágenes de Street View con texto descriptivo en una página HTML.
Ejemplo de flujo de trabajo para crear un recorrido:
Obtener imágenes de diferentes ubicaciones:
get_street_view(filename="empire.jpg", location="Empire State Building, NY")
get_street_view(filename="times_square.jpg", location="Times Square, NY")
get_street_view(filename="central_park.jpg", location="Central Park, NY")Crear una página de recorrido HTML:
create_html_page(
filename="nyc_tour.html",
title="New York City Tour",
html_elements=[
"<h1>New York City Landmarks Tour</h1>",
"<p>Explore these famous NYC landmarks through Street View images.</p>",
"<h2>Empire State Building</h2>",
"<img src='../output/empire.jpg' alt='Empire State Building'>",
"<p class='location'>350 Fifth Avenue, New York, NY</p>",
"<p class='description'>An iconic 102-story Art Deco skyscraper in Midtown Manhattan.</p>",
"<h2>Times Square</h2>",
"<img src='../output/times_square.jpg' alt='Times Square'>",
"<p class='location'>Broadway & 7th Avenue, New York, NY</p>",
"<p class='description'>Famous for its bright lights, Broadway theaters, and as the site of the annual New Year's Eve ball drop.</p>",
"<h2>Central Park</h2>",
"<img src='../output/central_park.jpg' alt='Central Park'>",
"<p class='location'>Central Park, New York, NY</p>",
"<p class='description'>An urban park spanning 843 acres in the heart of Manhattan.</p>"
]
)Estructura del proyecto
street_view_mcp/__init__.py: Inicialización del paquetemain.py: Punto de entrada para el servidor MCPserver.py: implementación del servidor MCPstreet_view.py: Cliente de la API principal de Street View
Notas importantes
Almacenamiento local : esta herramienta guarda todas las imágenes de Street View y los archivos HTML localmente en el directorio
output/Sin limpieza automática : no hay un mecanismo integrado para eliminar archivos guardados
Limpieza manual : debe limpiar periódicamente el directorio
output/para administrar el espacio en discoUso de API : cada solicitud de imagen cuenta para su cuota de API de Google Maps y puede generar cargos.
Desarrollo
Pruebas
pytestLicencia
Instituto Tecnológico de Massachusetts (MIT)
Available Tools
4 toolscreate_html_pageA
Create an HTML page specifically for displaying Street View images with descriptive text.
This tool is designed to compile multiple Street View images into a single viewable HTML document, creating a virtual tour or location showcase. The function automatically wraps your content in a complete HTML document with:
DOCTYPE declaration
HTML, head, and body tags
Basic responsive styling optimized for displaying images
Title from the parameter
Args: html_elements: List of content HTML elements (just the body content, no need for HTML structure) filename: Name of the HTML file to create (without directory path) title: Title for the HTML page
Returns: Dict: A status message indicating success or failure
Raises: ValueError: If the filename already exists or is invalid
Note:
- You only need to provide the CONTENT elements (no need for html, head, body tags)
- IMPORTANT: When including Street View images, you MUST use the path "../output/":
<img src="../output/empire.jpg" alt="Empire State Building">
- The "../" prefix is REQUIRED because HTML files are in html/ directory while
images are in output/ directory (both at the same level)
Example usage: ``` # Create a virtual Street View tour with multiple locations html_elements = [ "New York City Landmarks Tour", "Explore famous landmarks through Street View images.",
"<h2>Empire State Building</h2>",
"<img src='../output/empire.jpg' alt='Empire State Building'>",
"<p class='location'>350 Fifth Avenue, New York, NY</p>",
"<p class='description'>This 102-story Art Deco skyscraper in Midtown Manhattan was completed in 1931.</p>",
"<h2>Times Square</h2>",
"<img src='../output/timessquare.jpg' alt='Times Square'>",
"<p class='location'>Broadway & 7th Avenue, New York, NY</p>",
"<p class='description'>Famous for its bright lights, Broadway theaters, and as the site of the annual New Year's Eve ball drop.</p>"
]
```HTML Boilerplate (automatically added):
<!DOCTYPE html> <html lang="en"> <head> <meta charset="UTF-8"> <meta name="viewport" content="width=device-width, initial-scale=1.0"> <title>{title}</title> <style> body { font-family: Arial, sans-serif; line-height: 1.6; max-width: 800px; margin: 0 auto; padding: 20px; color: #333; } img { max-width: 100%; height: auto; border-radius: 5px; margin: 20px 0; } h1, h2, h3 { color: #2c3e50; } </style> </head> <body> <!-- Your content elements are inserted here --> </body> </html>
Inputs: html_elements (required), filename (required), title.
get_metadataB
Fetch metadata about a Street View panorama.
Args: location: The address to check for Street View imagery lat_lng: Comma-separated latitude and longitude (e.g., "40.748817,-73.985428") pano_id: Specific panorama ID to fetch metadata for radius: Search radius in meters when using location or coordinates source: Limit Street View searches to selected sources ("default" or "outdoor")
Returns: Dict: Panorama metadata including status, copyright, date, pano_id, lat, lng
Inputs: location, lat_lng, pano_id, radius, +1 more.
get_street_viewA
Fetch a Street View image based on location, coordinates, or panorama ID and save to file.
Args: filename: Required filename to save the image (must not already exist in output directory) location: The address to get Street View image for (e.g., "Empire State Building, NY") lat_lng: Comma-separated latitude and longitude (e.g., "40.748817,-73.985428") pano_id: Specific panorama ID to fetch size: Image dimensions as "widthxheight" (e.g., "600x400") heading: Camera heading in degrees (0-360) pitch: Camera pitch in degrees (-90 to 90) fov: Field of view in degrees (zoom level, 10-120) radius: Search radius in meters when using location or coordinates source: Limit Street View searches to selected sources ("default" or "outdoor")
Returns: Image: The Street View image
Raises: ValueError: If filename already exists in output directory
Inputs: filename (required), location, lat_lng, pano_id, +6 more.
open_image_locallyB
Open a saved Street View image in the default application.
Args: filename: The filename of the image to open (must exist in output directory)
Returns: Dict: A status message indicating success or failure
Raises: ValueError: If the file doesn't exist in the output directory
Inputs: filename (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
Each tool has a distinct, non-overlapping purpose: get_metadata retrieves panorama data, get_street_view fetches and saves images, create_html_page compiles images into HTML documents, and open_image_locally opens saved images. The boundaries are clear with no ambiguity between tools.
All tool names follow a consistent verb_noun pattern: get_metadata, get_street_view, create_html_page, open_image_locally. The naming is uniform and predictable, using clear action verbs followed by descriptive nouns.
With 4 tools, this server is well-scoped for Street View functionality. Each tool serves a specific, necessary role in the workflow: metadata retrieval, image acquisition, HTML compilation, and local viewing. No tool feels redundant or missing for the domain.
The toolset covers the core Street View workflow comprehensively: fetching metadata, acquiring images, creating HTML tours, and viewing images locally. A minor gap exists in lacking direct image manipulation or advanced HTML customization tools, but the basic lifecycle is fully covered without dead ends.
Maintenance
Related MCP Connectors
The Google Maps MCP server is a fully-managed server provided by the Maps Grounding Lite API that connects AI applications to Google Maps Platform services. It provides three main tools for building LLM applications: searching for places, looking up weather information, and computing routes with details like distance and travel time. The server acts as a proxy that translates Google Maps data into a format that AI applications can understand, enabling agents to accurately answer real-world location and travel queries.
Create immersive travel experiences by instructing an avatar to navigate Google Maps. Report on th…
Geospatial AI MCP server — satellite imagery, embeddings, weather, GNS governance
Ground your AI applications with trusted geospatial data from Google Maps.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceA server that provides AI-powered image generation, modification, and processing capabilities through the Model Context Protocol, leveraging Google Gemini models and other image services.19MIT
- AlicenseBqualityBmaintenanceA Model Context Protocol server that provides Google Maps API integration, allowing users to search locations, get place details, geocode addresses, calculate distances, obtain directions, and retrieve elevation data through LLM processing capabilities.71,362 npm502MIT
- AlicenseAqualityDmaintenanceAn AI-powered server that helps users discover and book restaurants based on location, cuisine preferences, mood, and event type, with integration to Google Maps Places API for accurate recommendations.516MIT
- AlicenseNot gradedqualityNot gradedmaintenanceA server that enables AI assistants to control a browser through tools, allowing them to perform web automation tasks like navigation, typing, clicking, and taking screenshots.Apache 2.0