Banking Chatbot with MCP Integration

A sophisticated banking chatbot application that uses Azure OpenAI and Model Context Protocol (MCP) for secure and efficient message handling.

Features

• AI-Powered Banking Assistant: Uses Azure OpenAI to provide intelligent responses to banking queries

• Model Context Protocol (MCP): Implements a secure message communication protocol

• Real-time Chat Interface: Modern, responsive UI for seamless user interaction

• Comprehensive Logging: Detailed logging system for monitoring and debugging

• Bank Information Integration: Dynamic display of bank details and services

• Markdown Support: Rich text formatting for responses

Related MCP server: agentbox

Project Structure

.
├── app.py                  # Main Flask application
├── mcp_server.py          # MCP server implementation
├── mcp_client.py          # MCP client implementation
├── requirements.txt       # Python dependencies
├── .env                   # Environment variables
├── templates/             # HTML templates
│   └── index.html        # Chat interface
└── logs/                 # Log files
    ├── client_messages.log
    ├── mcp_client.log
    └── mcp_server.log

Prerequisites

• Python 3.8 or higher

• Azure OpenAI API access

• Required Python packages (see requirements.txt)

Installation

1. Clone the repository:

   git clone <repository-url>
   cd banking-chatbot

2. Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Create a .env file with your credentials:

   ENDPOINT_URL=your_azure_endpoint
   AZURE_OPENAI_API_KEY=your_api_key
   DEPLOYMENT_NAME=your_deployment_name

Usage

1. Start the MCP server:

   python mcp_server.py

2. In a new terminal, start the Flask application:

   python app.py

3. Access the chatbot interface at http://localhost:5000

MCP Protocol

The Model Context Protocol (MCP) is implemented to handle message communication between the chatbot and the server. It provides:

• Secure message transmission

• Message queuing and reliability

• Detailed logging

• Real-time message handling

Message Types

• Chat Messages: User queries and AI responses

• System Messages: Administrative and control messages

Logging

The application maintains detailed logs in the logs directory:

• client_messages.log: Chat message history

• mcp_client.log: Client connection and operation logs

• mcp_server.log: Server operation logs

Bank Information

The chatbot is configured with comprehensive bank information including:

• Business hours

• Branch locations

• Available services

• Contact information

• Support channels

Development

Adding New Features

1. Update the BANK_INFO dictionary in app.py for new bank information

2. Modify the SYSTEM_MESSAGE for updated AI behavior

3. Add new message handlers in mcp_client.py for additional functionality

Testing

Run the test client to verify MCP functionality:

python test_client.py

Clear logs for testing:

python clear_logs.py

Security

• API keys and sensitive information are stored in .env

• MCP provides secure message transmission

• Input validation and error handling are implemented

Contributing

1. Fork the repository

2. Create a feature branch

3. Commit your changes

4. Push to the branch

5. Create a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.