get_forecast
Get 5-day weather forecasts for Portuguese cities including temperature, precipitation, wind conditions, and weather types to help plan activities and prepare for changing conditions.
Instructions
Get 5-day weather forecast for a Portuguese city (Previsão Meteorológica até 5 dias).
Args:
city_name: Name of the Portuguese city (e.g. Lisboa, Porto, Faro, Aveiro, Braga)
Returns detailed 5-day forecast including temperature, precipitation, wind, and weather type.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| city_name | Yes |
Implementation Reference
- weather.py:59-115 (handler)The get_forecast tool handler, registered via @mcp.tool() decorator. It retrieves a 5-day weather forecast for a specified Portuguese city by querying the IPMA API for city ID and forecast data, then formats and returns the information as a string.@mcp.tool() async def get_forecast(city_name: str) -> str: """Get 5-day weather forecast for a Portuguese city (Previsão Meteorológica até 5 dias). Args: city_name: Name of the Portuguese city (e.g. Lisboa, Porto, Faro, Aveiro, Braga) Returns detailed 5-day forecast including temperature, precipitation, wind, and weather type. """ # First, get the global ID for the city cities_url = f"{IPMA_API_BASE}/distrits-islands.json" cities_data = await make_ipma_request(cities_url) if not cities_data or "data" not in cities_data: return "Unable to fetch city database." # Search for the best match city_name_lower = city_name.lower() best_match = None for entry in cities_data["data"]: local_name = entry.get("local", "").lower() if city_name_lower in local_name or local_name in city_name_lower: best_match = entry break if not best_match: # Return available cities if no match found available_cities = [entry.get("local", "") for entry in cities_data["data"][:15]] return f"City '{city_name}' not found. Some available cities: {', '.join(available_cities)}" # Get the forecast using the global ID global_id = best_match.get("globalIdLocal") forecast_url = f"{IPMA_API_BASE}/forecast/meteorology/cities/daily/{global_id}.json" forecast_data = await make_ipma_request(forecast_url) if not forecast_data or "data" not in forecast_data: return "Unable to fetch forecast data for this location." # Format the forecast data city_info = f"5-Day Weather Forecast for {best_match.get('local')}\n" city_info += f"Location: {best_match.get('latitude')}°N, {best_match.get('longitude')}°E\n\n" forecasts = [] for day in forecast_data["data"][:5]: # Show next 5 days forecast = f"""Date: {day.get('forecastDate', 'Unknown')} Max Temperature: {day.get('tMax', 'N/A')}°C Min Temperature: {day.get('tMin', 'N/A')}°C Precipitation Probability: {day.get('precipitaProb', 'N/A')}% Weather Type ID: {day.get('idWeatherType', 'N/A')} Wind Direction: {day.get('predWindDir', 'N/A')} Wind Speed Class: {day.get('classWindSpeed', 'N/A')} Precipitation Intensity Class: {day.get('classPrecInt', 'N/A')} """ forecasts.append(forecast) return city_info + "---\n".join(forecasts)
- weather.py:12-20 (helper)Helper function used by get_forecast to make HTTP requests to the IPMA API with error handling.async def make_ipma_request(url: str) -> dict[str, Any] | list[dict[str, Any]] | None: """Make a request to the IPMA API with proper error handling.""" async with httpx.AsyncClient() as client: try: response = await client.get(url, timeout=30.0) response.raise_for_status() return response.json() except Exception: return None
- weather.py:60-67 (schema)Input schema (city_name: str) and output type (str) defined in function signature and docstring.async def get_forecast(city_name: str) -> str: """Get 5-day weather forecast for a Portuguese city (Previsão Meteorológica até 5 dias). Args: city_name: Name of the Portuguese city (e.g. Lisboa, Porto, Faro, Aveiro, Braga) Returns detailed 5-day forecast including temperature, precipitation, wind, and weather type. """
- weather.py:59-59 (registration)The @mcp.tool() decorator registers the get_forecast function as an MCP tool.@mcp.tool()