Adithya-R35/mcp-email-agent

MCP Database Server with OAuth Email Agent

An intelligent email-based data access system that combines a Model Control Protocol (MCP) server for database queries with an AI-powered email agent using Gmail OAuth2 authentication.

Features

MCP Database Server (mcp_server.py)

• MSSQL Integration: Direct connection to Microsoft SQL Server databases
• Resource Management: Pre-defined database resources for common queries
• Health Monitoring: Built-in connection health checks
• Flexible Configuration: Environment-based configuration support

OAuth Email Agent (oauth_email_agent.py)

• Gmail OAuth2: Secure authentication with Gmail API
• AI-Powered Processing: Uses OpenAI GPT-4 for intelligent email parsing
• Authorization System: Role-based access control for data requests
• CSV Export: Automatic data export with email attachments
• LangGraph Workflow: Structured processing pipeline

Available Data Resources

1. Customer Data - Complete customer information and contact details
2. Product Catalog - Product details, inventory, and pricing
3. Customer Addresses - Address information and locations
4. Sales Order Details - Transaction history and order data
5. Product Descriptions - Detailed product specifications

Quick Start

1. Clone Repository

git clone https://github.com/yourusername/mcp-email-agent.git
cd mcp-email-agent

2. Install Dependencies

pip install -r requirements.txt

3. Environment Setup

cp .env.example .env
# Edit .env with your database and API credentials

4. Gmail OAuth Setup

1. Go to Google Cloud Console
2. Create a new project or select existing
3. Enable Gmail API
4. Create OAuth 2.0 credentials
5. Download credentials.json to project root

5. Run the Services

Start MCP Server:

python mcp_server.py

Start Email Agent:

python oauth_email_agent.py

Configuration

Database Configuration

Set these environment variables in .env:

MSSQL_SERVER=your_server
MSSQL_DATABASE=your_database
MSSQL_USERNAME=your_username
MSSQL_PASSWORD=your_password

API Keys

OPENAI_API_KEY=your_openai_key

Usage

Email Requests

Send emails requesting data to the configured Gmail account:

Example requests:

• "I need customer contact information"
• "Can you send me the product inventory?"
• "Show me recent sales data"
• "I want customer addresses for shipping"

Authorization

Configure authorized users in oauth_email_agent.py:

authorized_clients = [
    AuthorizedClient(
        email="user@company.com",
        name="John Doe",
        company="Your Company",
        authorized_resources=["*"]  # or specific resources
    )
]

Response Types

• ✅ Data Response: CSV attachment with requested data
• ❌ Unauthorized: Access denied message
• ❓ Clarification: Request for more specific information
• 🚫 No Data: Resource not available message

Security Features

• OAuth2 authentication for Gmail access
• Role-based authorization system
• Resource-level access control
• Secure credential management
• No hardcoded passwords or tokens

Architecture

Email Request → Gmail API → LangGraph Workflow → MCP Server → Database
                                ↓
CSV Response ← Gmail API ← AI Response Generation ← Query Results

Dependencies

• Database: pyodbc for MSSQL connectivity
• AI: langchain-openai for GPT-4 integration
• Workflow: langgraph for process orchestration
• Gmail: google-api-python-client for OAuth2 and API access
• MCP: mcp for Model Control Protocol server

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

Security Notes

• Never commit credentials.json or token.json
• Keep .env files out of version control
• Use strong database passwords
• Regularly rotate API keys
• Monitor authorized user list

License

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

Support

For issues and questions:

1. Check the GitHub Issues page
2. Review the setup documentation in /setup/
3. Ensure all dependencies are properly installed
4. Verify environment variables are correctly set

Roadmap

• Support for additional database types (PostgreSQL, MySQL)
• Web interface for authorization management
• Advanced query filtering and parameterization
• Slack integration
• Enhanced logging and monitoring
• Multi-language support