MCP Demo

An MCP demonstration project using the DeepSeek-V3 model to query weather information through command-line interaction, supporting functions to obtain alerts and weather forecasts.

Artificial intelligence chatbots Location services #Weather query #Command-line tool #LLM application .Python

rating : 2 points

downloads : 11

update time : 2025-04-29

What is MCP Demo Server?

The MCP Demo Server is a practical implementation of the Model Context Protocol that connects language models with external tools. This specific demo provides weather forecasting capabilities by integrating with the National Weather Service API.

How to use the MCP Demo Server?

You can interact with the server through a simple client interface that sends natural language queries about weather conditions. The system will automatically determine the appropriate weather data to retrieve and present it in an easy-to-understand format.

Use Cases

Ideal for checking current weather conditions, planning outdoor activities, or integrating weather data into other applications through natural language queries.

Key Features

Weather ForecastProvides detailed weather forecasts including temperature, wind conditions, and sky conditions for locations worldwide

Weather AlertsCan check for active weather alerts and warnings in specified locations

Natural Language InterfaceUnderstands and responds to natural language queries about weather conditions

Pros and Cons

Advantages

Easy-to-use natural language interface

Provides comprehensive weather information from reliable sources

Can handle complex queries about weather conditions

Quick response time for weather data

Limitations

Limited to weather-related queries only

Requires an API key for DeepSeek integration

Accuracy depends on the underlying weather data sources

Currently only supports English queries

Getting Started

Set up your environment

Install the required dependencies and create a virtual environment

Install required packages

Add the necessary Python packages including MCP and its dependencies

Configure API key

Add your DeepSeek API key to the .env file

Run the client

Start interacting with the weather service through the client interface

Example Queries

Basic Weather QueryGetting the current weather forecast for a major city

Multi-day ForecastRequesting weather information for multiple days

Weather Alerts CheckChecking for severe weather alerts

Frequently Asked Questions

What weather data sources does this use?

Do I need a DeepSeek API key?

Can I use this with other LLMs besides DeepSeek?

What locations are supported?

How accurate is the weather data?

Additional Resources

MCP Documentation

Official documentation for Model Context Protocol

National Weather Service API

Documentation for the weather data API used by this demo

DeepSeek API

Information about the DeepSeek language model API

Example Repository

Source code for this demonstration project

🚀 MCP Demo Project

This project provides a simple way to interact with weather forecasts using the MCP framework. It demonstrates how to set up the environment, run the client, and get weather information.

🚀 Quick Start

📦 Installation

To install the project, follow these steps:

uv init mcp-demo
cd mcp-demo
uv venv
source .venv/bin/activate
uv add "mcp[cli]" httpx
uv add mcp openai python-dotenv
rm main.py

💻 Usage Examples

Basic Usage

Before running the client, replace the OPENAI_API_KEY in the .env file with your deepseek api key. Then, run the following command to start the client:

python3 client.py weather.py

Output Example

Here is an example of the output you might see when running the client:

[04/23/25 21:11:54] INFO     Processing request of type ListToolsRequest                                                                                 server.py:534

Connected to server with tools: ['get_alerts', 'get_forecast']

MCP client has been started!
Enter your query or type 'quit' to exit.

Query: What's the weather like in New York?
[04/23/25 21:12:09] INFO     Processing request of type ListToolsRequest                                                                                 server.py:534
[04/23/25 21:12:15] INFO     Processing request of type CallToolRequest                                                                                  server.py:534
                    INFO     HTTP Request: GET https://api.weather.gov/points/40.7128,-74.006 "HTTP/1.1 200 OK"                                        _client.py:1740
[04/23/25 21:12:16] INFO     HTTP Request: GET https://api.weather.gov/gridpoints/OKX/33,35/forecast "HTTP/1.1 200 OK"                                 _client.py:1740
[DEBUG] final_text[0] type: <class 'str'>, value: The current weather forecast for New York is as follows:

### Today:
- **Temperature**: 73°F  
- **Wind Speed**: 7 mph NW  
- **Forecast**: Mostly sunny, with a high near 73.

### Tonight:
- **Temperature**: 57°F  
- **Wind Speed**: 3 to 7 mph SE  
- **Forecast**: Mostly clear.

### Tomorrow:
- **Temperature**: 68°F (falling to 64°F in the afternoon)  
- **Wind Speed**: 3 to 12 mph SE  
- **Forecast**: Sunny.

### Tomorrow night:
- **Temperature**: 56°F  
- **Wind Speed**: 6 to 12 mph S  
- **Forecast**: Mostly clear.

### The day after tomorrow:
- **Temperature**: 70°F (falling to 66°F in the afternoon)  
- **Wind Speed**: 5 to 14 mph S  
- **Forecast**: Partly cloudy.

If you need more details, let me know!

The current weather forecast for New York is as follows:

### Today:
- **Temperature**: 73°F  
- **Wind Speed**: 7 mph NW  
- **Forecast**: Mostly sunny, with a high near 73.

### Tonight:
- **Temperature**: 57°F  
- **Wind Speed**: 3 to 7 mph SE  
- **Forecast**: Mostly clear.

### Tomorrow:
- **Temperature**: 68°F (falling to 64°F in the afternoon)  
- **Wind Speed**: 3 to 12 mph SE  
- **Forecast**: Sunny.

### Tomorrow night:
- **Temperature**: 56°F  
- **Wind Speed**: 6 to 12 mph S  
- **Forecast**: Mostly clear.

### The day after tomorrow:
- **Temperature**: 70°F (falling to 66°F in the afternoon)  
- **Wind Speed**: 5 to 14 mph S  
- **Forecast**: Partly cloudy.

If you need more details, let me know!

Query:

⚠️ Important Note

Make sure to replace the OPENAI_API_KEY in the .env file with your valid deepseek api key before running the client. Otherwise, the client may not function properly.

💡 Usage Tip

You can modify the client.py and weather.py files to customize the functionality of the client, such as adding more tools or changing the query format.