MCP_TERMINAL_SERVER

gabriellemosc/MCP_TERMINAL_SERVER

3.1

If you are the rightful owner of MCP_TERMINAL_SERVER and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

The Model Context Protocol (MCP) server facilitates communication between a virtual assistant and a database, enabling efficient data retrieval and interaction.

Virtual Car Agent

An interactive virtual assistant developed in Python to help users find cars based on filters such as make, model, year, fuel type, transmission, price, and more. The agent chats with the user, searches a database via the MCP protocol, and returns results in a clear and user-friendly format.

Project Structure

Tech-Challege-C2S/
│
├─ agent.py               # Main assistant code
├─ carros.csv             # Sample vehicle data
├─ inventario.db          # SQLite database (not versioned)
├─ models.py              # Database models
├─ popular_bd.py          # Script to populate the database
├─ requirements.txt       # Project dependencies
├─ README.md              # Project documentation
├─ venv/                  # Virtual environment (not versioned)
└─ mcp/                   # MCP protocol module
   ├─ __init__.py         # Package initializer
   ├─ client.py           # MCP client
   ├─ server.py           # MCP server
   ├─ message_type.py     # Protocol message types
   └─ protocol.py         # MCP protocol logic

Installation & Setup

  1. Clone this repository:
git clone https://github.com/gabriellemosc/C2S---Challenge.git
cd C2S---Challenge
  1. Create and activate a virtual environment (optional but recommended):
python3 -m venv venv
source venv/bin/activate   # Linux / macOS
venv\Scripts\activate      # Windows
  1. Install the dependencies
pip install -r requirements.txt
  1. Since the database (inventario.db) is not included, you need to create and populate it
  • 4.1 Run the models to create tables: 
     python3 models.py
  • 4.2 Populate the database with sample data: 
    
python3 popular_bd.py

This script will populate the SQLite database (inventario.db) with random vehicles for testing purposes.

  1. Start the MCP server in a separate terminal:
python3 -m mcp.server

How to Use

  1. Run the agent in a new terminal (keep the server running):
python3 agent.py

  1. Follow the assistant instructions:

    • Enter the car’s make, model, year, fuel type, and price range.
    • The agent will query the database via MCP and display the results.
    • You can choose to perform a new search or exit at the end.

  2. Example Interaction:

👋 Hello! I'm Gabriel, your virtual assistant to find your car.
Let's talk about what you're looking for...

Ok! Which car make are you interested in? Toyota, Honda, Volkswagen, etc...
> Toyota

And within Toyota, do you have a preferred model?
> Corolla

What about the year? Any specific range?
> 2018-2022

...
🔎 Alright, let me check what we have.
This might take a moment...

✅ I found 3 cars that might interest you:
1. 🟢 Toyota Corolla (2019)
     Black | 📊 35,000 km
     ⚙️ Transmission: Automatic | ⛽ Fuel: Flex
     💰 $95,000.00
     ✅ Available for test drive!

🔁 Do you want to make a new search? (y/n):

MCP Protocol

  • The MCP module handles communication between the agent and the database server.
  • Server: mcp.server
  • Client: mcp.client
  • Protocol logic: mcp.protocol
  • Message types: mcp.message_type

Make sure the server is running before starting the agent.

Contact

Gabriel Lemos