sap-datasphere-mcp-server by MarioDeFelipe - MCP Server

SAP Datasphere MCP Server

A Model Context Protocol (MCP) server that provides AI assistants with seamless access to SAP Datasphere capabilities including space management, data discovery, and analytics operations.

🚀 Quick Start

# Clone the repository
git clone https://github.com/yourusername/sap-datasphere-mcp-server.git
cd sap-datasphere-mcp-server

# Install dependencies
pip install mcp pydantic requests

# Test the server
python test_simple_server.py

✨ Features

🏢 Space Management

List all Datasphere spaces with status and metadata
Get detailed space information including tables and views
Explore space configurations and permissions

🔍 Data Discovery

Search tables and views across all spaces
Get detailed table schemas with column information
Explore data catalog and metadata

🔗 Connection Management

List and monitor data source connections
Check connection status and health
View connection configurations

📊 Data Querying

Execute SQL queries against Datasphere data
View query results and performance metrics
Simulate analytical operations

🛒 Marketplace Integration

Browse available data packages
Search marketplace content by category
View package details and installation options

🎯 MCP Tools Available

Tool	Description	Example Usage
`list_spaces`	List all Datasphere spaces	"Show me all available spaces"
`get_space_info`	Get detailed space information	"Tell me about the Sales Analytics space"
`search_tables`	Search for tables across spaces	"Find tables containing customer data"
`list_connections`	List data source connections	"What data connections are available?"
`execute_query`	Execute SQL queries	"Run: SELECT * FROM CUSTOMER_DATA LIMIT 10"

📋 MCP Resources

datasphere://spaces - Access to all Datasphere spaces
datasphere://connections - Data source connection information

⚙️ Configuration

Claude Desktop Setup

Add to your Claude Desktop MCP configuration:

{
  "mcpServers": {
    "sap-datasphere": {
      "command": "python",
      "args": ["path/to/sap_datasphere_mcp_simple.py"],
      "env": {}
    }
  }
}

Mock vs Live Mode

The server supports two modes:

Mock Mode (Default): Uses realistic sample data for development and testing

2 sample spaces (Sales Analytics, Finance DWH)
Sample tables with proper schemas
Simulated connections and query results

Live Mode: Connects to real SAP Datasphere APIs (requires OAuth2 setup)

🔧 Installation

Prerequisites

Python 3.10 or higher
MCP-compatible AI assistant (Claude Desktop, etc.)

Dependencies

pip install mcp>=1.0.0 pydantic>=2.0.0 requests>=2.31.0

Optional: SAP HANA Client

For direct database access:

pip install hdbcli>=2.26.0

💡 Usage Examples

Basic Queries

"List all Datasphere spaces"
"Show me tables in the Sales Analytics space"
"What's the schema of the CUSTOMER_DATA table?"
"List all data connections and their status"

Advanced Operations

"Search for all tables containing sales data"
"Execute: SELECT COUNT(*) FROM CUSTOMER_DATA"
"Show me connection details for SAP ERP"
"What data packages are available in the marketplace?"

🔐 OAuth2 Setup (For Live Mode)

To connect to real SAP Datasphere APIs:

Get OAuth Credentials from your Datasphere administrator

Update Configuration in the server file:

DATASPHERE_CONFIG = {
    "use_mock_data": False,
    "oauth_config": {
        "client_id": "your-client-id",
        "client_secret": "your-client-secret",
        "token_url": "https://your-tenant.eu10.hcs.cloud.sap/oauth/token"
    }
}

Restart the MCP server

🧪 Testing

Run the test suite to validate functionality:

python test_simple_server.py

Expected output:

✅ Mock data validation
✅ MCP configuration generation
✅ Usage examples and setup instructions

🏗️ Architecture

The MCP server follows clean architecture principles:

Mock Data Layer: Realistic sample data for development
MCP Protocol Layer: Standard MCP tools and resources
SAP Integration Layer: OAuth2 and API connectivity (ready for live mode)
Error Handling: Comprehensive error handling and logging

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Fork the repository
Create your feature branch (git checkout -b feature/AmazingFeature)
Commit your changes (git commit -m 'Add some AmazingFeature')
Push to the branch (git push origin feature/AmazingFeature)
Open a Pull Request

📝 License

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

🆘 Support

📋 Open an issue
📚 Check SAP Datasphere documentation
💬 Join the discussion in Issues

🙏 Acknowledgments

SAP Datasphere team for the excellent platform
Model Context Protocol community
Contributors and users of this project

Ready to explore your SAP Datasphere data with AI assistants! 🚀