MCP LangChain Weather Agent
A simple weather MCP server connected to a LangGraph agent. The agent calls the weather tool to fetch live weather for any city.
Setup
- Install dependencies:
pip install -r requirements.txt- Create a
.envfile in the project root:
OPENROUTER_API_KEY=your_api_key_hereUsage
Start the MCP server (runs on http://127.0.0.1:8000):
python weatherserver.pyIn a second terminal, run the agent client:
python client.pyThe client connects to the weather MCP server, loads the weather tool, and asks the LLM about the weather.
Files
| File | Description | |------|-------------| | weatherserver.py | MCP server with a weather tool (geocoding + Open-Meteo API) | | client.py | LangGraph agent that uses the MCP weather tool | | requirements.txt | Python dependencies |
How it works
- User asks a question like "what's the weather in Tokyo?"
- The agent calls the
weatherMCP tool - The server geocodes the city name and fetches current weather from Open-Meteo
- The agent returns the result to the user
