# MeteoSource MCP > MeteoSource connects your AI agent to global weather data. Use it to search for specific locations or find the nearest monitored spot using `get_nearest_weather_place`. You can then audit detailed point forecasts, check timezones with `get_place_timezone`, and manage complex location metadata by searching places first with `search_weather_places` before running a forecast. ## Overview - **Category:** data-analytics - **Price:** Free - **Tags:** weather-forecast, hyper-local-data, meteorology, environmental-data, api-integration, real-time-data ## Description MeteoSource plugs your AI agent straight into global weather data. You don't have to copy and paste from a dozen different websites. This server lets your agent run an entire meteorological workflow—from locating a place ID all the way through to generating hourly forecasts—all within the chat window. It handles the complicated parts so you don't have to. Whether you start with coordinates or just a city name, your agent figures out what it needs. You can check time zones for specific places, pinpoint the closest monitored weather station, and pull detailed daily or hourly forecasts immediately. First off, always run `check_api_status`. This tells you if the whole MeteoSource connection is online and actually working before you waste a shot trying to get data. If it's good to go, you can start finding locations. If you know what city you're looking for but don't have an official ID, run `search_weather_places`. This searches global spots by name and spits out the unique place IDs required for every other function. If you only have latitude and longitude—say, a GPS coordinate from a field worker—you use `get_nearest_weather_place`. It takes those coordinates and identifies the closest monitored weather station nearby. That tool gives you the ID of the spot you need to track. Once you've got that place ID, you can build out your context. Need to know if an event is happening at sunrise in a different time zone? Run `get_place_timezone` with the place ID; it returns all the associated timezone metadata, like 'Asia/Tokyo'. This lets you synchronize events across multiple global locations. Finally, when you need the weather itself, you use `get_point_forecast`. Just hand it that specific place ID. The tool pulls everything: current conditions summaries, full daily forecasts detailing high and low temperatures, or minute-by-minute hourly breakdowns. You get a complete picture of what's going down at that location. Your agent uses these tools together—searching first, grabbing the timezone context second, then pulling the forecast third—all without you needing to know which ID belongs where. It’s clean data access for complex field operations. ## Tools ### check_api_status Checks if the MeteoSource service is currently operational and ready for use. ### get_point_forecast Fetches detailed weather forecasts (current, daily, hourly) using a specific place ID. ### get_nearest_weather_place Finds and returns the closest monitored weather station given a latitude and longitude. ### get_place_timezone Retrieves timezone information for any specific, known place ID. ### search_weather_places Searches for global locations by name and returns the necessary place IDs to run other tools. ## Prompt Examples **Prompt:** ``` Get weather forecast for 'London' using MeteoSource. ``` **Response:** ``` I've retrieved the weather for London! The current temperature is 12°C with light rain. The daily summary shows similar conditions for the next 3 days. Would you like the detailed hourly breakdown? ``` **Prompt:** ``` Search for weather station near latitude 48.8566 and longitude 2.3522. ``` **Response:** ``` I've identified the nearest place! It corresponds to Paris, France. I can retrieve the full point forecast or timezone metadata for this location if you'd like. ``` **Prompt:** ``` What is the timezone for place 'tokyo'? ``` **Response:** ``` I've retrieved the timezone information. Tokyo is located in the 'Asia/Tokyo' timezone, currently identified as GMT+9. I can assist you with more temporal auditing if needed. ``` ## Capabilities ### Find locations by name The agent searches a place name and returns a unique ID required for all other weather functions. ### Get the closest station You provide latitude/longitude, and the tool identifies the nearest monitored weather location. ### Check time zone data The agent takes a place ID and returns all associated timezone metadata (e.g., 'Asia/Tokyo'). ### Fetch detailed forecasts You give the tool a place ID, and it retrieves current conditions, daily summaries, or hourly breakdowns. ### Verify service status The agent checks if the entire MeteoSource connection is currently online and working. ## Use Cases ### Prepping for a cross-country conference. The coordinator needs to know if multiple cities will be impacted by rain. They ask the agent, which first uses `search_weather_places` to get IDs for Denver and Chicago. The agent then runs `get_point_forecast` on both sets of IDs, giving a side-by-side comparison so they can warn attendees before they even book flights. ### Coordinating an outdoor film shoot. The location scout is only given coordinates. They ask the agent to run `get_nearest_weather_place`. The agent finds the closest monitored station and then uses that place ID with `get_point_forecast` to get a detailed hourly breakdown, showing exactly when rain or wind might hit. ### Auditing global supply chain timing. The operations lead needs to know the time zone difference between two ports (e.g., Singapore and Seattle). They run `get_place_timezone` on both IDs, confirming the offset. Then they use that metadata to adjust delivery timelines for their team. ### Building a global event calendar. The planner needs to track weather data across 50 different locations. They run `search_weather_places` in batches, collect the place IDs, and then feed them into a loop that calls `get_point_forecast` for every single location, building an automated report. ## Benefits - You get real-time weather data without leaving your chat window. Instead of opening five different websites, your agent calls `get_point_forecast` and hands you the full JSON payload immediately. It’s all in one place. - Never worry about time zones again. Use `get_place_timezone` to instantly know if a location is GMT+9 or something else, which saves massive headaches for international event planning. - If you only have GPS coordinates, don't guess. Running `get_nearest_weather_place` accurately pinpoints the correct monitored station, ensuring your forecast data is precise and localized. - Planning a regional audit? Use `search_weather_places` first to gather thousands of place IDs, then feed those bulk results into `get_point_forecast` for comprehensive oversight. - The whole system stays reliable. Before running any job, use `check_api_status`. Knowing the API is up prevents your agent from failing mid-task and wasting time. ## How It Works The bottom line is: your agent handles the API calls and data parsing; you just talk to it normally. 1. Subscribe to the server and give your AI client your unique MeteoSource API Key. 2. Call a tool (like `search_weather_places`) with basic input, such as a city name or coordinates. 3. The agent returns structured data—a place ID, timezone info, or forecast JSON—which you use for the next step. ## Frequently Asked Questions **How do I get a place ID using MeteoSource?** Use `search_weather_places`. You just provide the name of the city or location, and the tool returns the necessary unique place IDs needed for all other forecast tools. **Is `get_nearest_weather_place` better than searching by name?** Yes. If you only have latitude/longitude coordinates, `get_nearest_weather_place` is the right tool because it finds the closest *monitored* station to your raw point, guaranteeing data accuracy. **Can I check multiple time zones with MeteoSource?** Yes. Run `get_place_timezone` for each location you care about. This lets your agent compile a list of all the required timezone metadata (like 'America/Los_Angeles') in one query. **What if I need to check the API status first?** Start by calling `check_api_status`. This simple tool verifies the entire MeteoSource connection is live. It's smart practice before running any expensive or time-sensitive queries. **When I use `get_point_forecast`, what should my agent do if the Place ID is invalid?** If `get_point_forecast` receives an invalid place ID, it will return a specific error code and no forecast data. Your client must implement basic input validation before calling this tool to ensure accuracy. **Are there rate limits when using multiple tools like `search_weather_places` and `get_nearest_weather_place`?** Yes, the MeteoSource API enforces usage quotas. We recommend grouping related searches into fewer calls rather than running them in rapid succession to avoid hitting rate limits. **How does the data returned by `get_point_forecast` structure daily and hourly information?** The forecast response is structured as a nested object containing both summary fields and an array for detailed breakdowns. You access the full temporal detail through the dedicated 'hourly' key within the main forecast body. **What kind of authentication setup does MeteoSource require when I connect via MCP?** MeteoSource requires a unique API Key for all connections. You must securely store this key and pass it to your AI client's environment variables before running any tool. **How do I find my MeteoSource API Key?** Sign up for a free account at [**meteosource.com**](https://www.meteosource.com/), and you will find your API Key in your dashboard. Copy and paste it below. **Does it support location search by name?** Yes. Use the `search_weather_places` tool providing the city or place name. Your agent will return the unique `place_id` required for forecasts. **Are hourly forecasts included?** Yes. The `get_point_forecast` tool retrieves comprehensive weather metadata including current conditions, daily summaries, and hourly breakdowns.