# rossshannon/Weekly-Weather-mcp [Health: Active]

**Category:** 🗺️ Location Services  
**Repository:** https://github.com/rossshannon/weekly-weather-mcp.git  
**GitHub Stars:** 10  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/rossshannon-weekly-weather-mcp

## Description
Weekly Weather MCP server which returns 7 full days of detailed weather forecasts anywhere in the world.

## Claude Desktop Quick Installation
Heuristic fallback — verify the package name and runner against the repository README before running it. Uses `npx` (confidence: low):

```json
"mcpServers": {
  "weekly-weather-mcp": {
    "command": "npx",
    "args": ["-y","rossshannon-weekly-weather-mcp"],
    "env": {
      "OPENWEATHER_API_KEY": ""
    }
  }
}
```

**Requires environment variables:** `OPENWEATHER_API_KEY` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What rossshannon/Weekly-Weather-mcp MCP server does

The rossshannon/Weekly-Weather-mcp MCP server exposes weather data through two MCP tools backed by OpenWeatherMap One Call API 3.0. It supports locations worldwide when OpenWeatherMap can resolve the supplied name. The server handles conversion from a place name to latitude and longitude, so callers do not need to provide coordinates directly.

`get_weather` returns current conditions, hourly data for the next 48 hours, and daily data for eight days: the current day plus the following seven days. Daily results include morning, afternoon, and evening data points, along with weather summaries, precipitation probabilities, temperature, humidity, wind speed, and other detailed values. `get_current_weather` returns a smaller response containing present conditions only, including temperature, feels-like temperature, weather condition, humidity, and wind.

## How it works

Both tools accept a `location` string, an optional `api_key`, and a `timezone_offset` expressed in hours. The offset defaults to UTC and controls the timezone used for returned times. Examples of accepted location formats include `Paris`, `New York`, `London,GB`, and `Portland,OR,US`. Adding a country or state code can reduce ambiguity. Unresolvable locations produce an error.

The rossshannon/Weekly-Weather-mcp MCP server reads the OpenWeatherMap key from `OPENWEATHER_API_KEY` when the tool call does not supply `api_key`. Results come from the external OpenWeatherMap service, so availability and request limits depend on that service and the account associated with the key.

## Setup and configuration

Clone the repository, create a Python virtual environment if desired, and install the dependencies listed in `requirements.txt`. The server is started with the `weather_mcp_server.py` script. For MCP client configuration, the README shows a Python command pointing to the script, with `OPENWEATHER_API_KEY` supplied in the client environment. A virtual environment can be used by pointing the client at its Python executable.

Claude Desktop and Cursor are listed as example MCP-supported clients. The API key can instead be passed in each tool call, allowing the process to run without setting the environment variable. OpenWeatherMap One Call API 3.0 access must be enabled for the account. The README states that the service offers 1,000 calls per day in its free tier, with additional usage subject to the provider's pricing and account limits.

## Tools and capabilities

- Query current conditions for a named location.
- Retrieve hourly forecasts covering the next 48 hours.
- Retrieve daily forecasts covering eight days.
- Inspect morning, afternoon, and evening values for each forecast day.
- Use precipitation probabilities and weather summaries for planning.
- Request times using a specified timezone offset.

## Limitations and notes

The server depends on a valid OpenWeatherMap API key and active One Call API 3.0 access. Newly created or newly activated keys may take time to work. Requests beyond the account limit can return HTTP 429 responses. Location quality depends on OpenWeatherMap geocoding; ambiguous names should include country or state information. The server provides forecast data and does not itself create calendar events or perform other follow-up actions.

_Full upstream README: https://allmcps.com/mcp/rossshannon-weekly-weather-mcp/readme_