openmcp-document/plugin-tutorial/examples/python-simple-stdio.md

# Implementing a Weather Information MCP Server in Python

[Video Tutorial](https://www.bilibili.com/video/BV1zYGozgEHc)

## The Problem Scenario

Before we begin, consider this common situation: I'm planning to attend the Arknights "Rusty Shadows" convention next Saturday in Hangzhou and want to check the weather forecast. When I ask an LLM about Saturday's weather, I get this response:

<div align=center>
<img src="https://picx.zhimg.com/80/v2-4c623ac6897e12093535b0d9ed9cf242_1440w.png" style="width: 100%;"/>
</div>

This "teach you to fish" approach isn't helpful for simple everyday queries. While there are many weather apps available, how can we integrate weather data directly into LLMs to get actionable answers?

## Introduction

👉 [Previous Guide](https://zhuanlan.zhihu.com/p/32593727614)

In our last tutorial, we covered MCP fundamentals. Now, we'll develop our first MCP server to bridge existing applications/services with LLMs.

For efficient development, we'll use OpenMCP - an integrated MCP testing tool I recently open-sourced:

[OpenMCP Announcement](https://zhuanlan.zhihu.com/p/1894785817186121106)  
OpenMCP GitHub: https://github.com/LSTM-Kirigaya/openmcp-client  
(Stars appreciated! :D)

### Initial Setup

First, install the UV tool (a Conda alternative):

```bash
pip install uv
uv  # Verify installation
```

Create a new project:
```bash
mkdir simple-mcp && cd simple-mcp
uv init
uv add mcp "mcp[cli]"
```

Install the OpenMCP plugin in VSCode:

<div align=center>
<img src="https://picx.zhimg.com/80/v2-525c4576398078547fdd6eeef26532aa_1440w.png" style="width: 100%;"/>
</div>

### Basic MCP Server

Create `simple_mcp.py`:

```python
from mcp.server.fastmcp import FastMCP

mcp = FastMCP('Weather MCP Server', version="1.0.0")

@mcp.tool(
    name='weather',
    description='Get weather information for specified city'
)
def get_weather(city: str) -> str:
    """Weather query protocol - returns formatted string"""
    return f"Weather in {city}: Sunny, 25°C"

# Additional example tools/resources omitted for brevity
```

Test the server:
```bash
uv run mcp run simple_mcp.py
```

### Connecting with OpenMCP

Click the purple OpenMCP icon in VSCode to launch the debugger. Verify connection status (green indicator):

<div align=center>
<img src="https://picx.zhimg.com/80/v2-c4ebbbfe98d51e8b6e7de6c6d1bceb2e_1440w.png" style="width: 100%;"/>
</div>

## Developing the Weather Function

### Tool Debugging

Our initial weather tool just returns static data. Let's test it in OpenMCP:

<div align=center>
<img src="https://picx.zhimg.com/80/v2-d32a9c0d9fcab497dc03152a72c4c62b_1440w.png" style="width: 100%;"/>
</div>

### Interactive Testing

Configure your LLM API in OpenMCP:

<div align=center>
<img src="https://pic1.zhimg.com/80/v2-367780b204d2aa50354585272b71af20_1440w.png" style="width: 100%;"/>
</div>

Test without tools:
```
What's the temperature in Hangzhou?
```

Then with our weather tool enabled:

<div align=center>
<img src="https://picx.zhimg.com/80/v2-e934d386e20b1de43fb5e0dd426de86e_1440w.png" style="width: 100%;"/>
</div>

Notice the two-step process:
1. LLM calls our weather tool with `{"city": "Hangzhou"}`
2. Our server responds with formatted weather data
3. LLM generates final answer

## Production-Ready Implementation

Here's a complete weather implementation using a real API:

```python
import requests
import json
from typing import NamedTuple, Optional
from mcp.server.fastmcp import FastMCP

class CityWeather(NamedTuple):
    city_name_en: str
    city_name_cn: str 
    city_code: str
    temp: str
    wd: str
    ws: str
    sd: str
    aqi: str
    weather: str

def get_city_weather(city_code: str) -> Optional[CityWeather]:
    """Get weather by city code"""
    try:
        url = f"http://d1.weather.com.cn/sk_2d/{city_code}.html"
        headers = {
            "User-Agent": "Mozilla/5.0...",
            "Host": "d1.weather.com.cn",
            "Referer": "http://www.weather.com.cn/"
        }
        
        response = requests.get(url, headers=headers)
        response.raise_for_status()
        
        content = response.text.encode('latin1').decode('unicode_escape')
        json_str = content[content.find("{"):]
        weather_data = json.loads(json_str)
        
        return CityWeather(
            city_name_en=weather_data.get("nameen", ""),
            city_name_cn=weather_data.get("cityname", "").encode('latin1').decode('utf-8'),
            city_code=weather_data.get("city", ""),
            temp=weather_data.get("temp", ""),
            wd=weather_data.get("wd", "").encode('latin1').decode('utf-8'),
            ws=weather_data.get("ws", "").encode('latin1').decode('utf-8'),
            sd=weather_data.get("sd", ""),
            aqi=weather_data.get("aqi", ""),
            weather=weather_data.get("weather", "").encode('latin1').decode('utf-8')
        )
    except Exception as e:
        print(f"Weather query failed: {str(e)}")
        return None

mcp = FastMCP('Weather MCP', version="1.0.0")

@mcp.tool(
    name='get_weather_by_city_code',
    description='Get weather by city code (integer)'
)
def get_weather_by_code(city_code: int) -> str:
    weather_data = get_city_weather(str(city_code))
    return str(weather_data)
```

Key Notes:
1. Use `int` type for numeric IDs to ensure proper JSON serialization
2. Follow Python naming conventions for tool names

### Final Test

<div align=center>
<img src="https://picx.zhimg.com/80/v2-e581c6461190b358adda50ce83633520_1440w.png" style="width: 100%;"/>
</div>

Success! We've built a fully functional weather MCP service. For production deployments, consider using SSE connections for better scalability (covered in future tutorials).

OpenMCP GitHub: https://github.com/LSTM-Kirigaya/openmcp-client