MCP Connector: Integrating AI agent with Data Warehouse in Microsoft Fabric
@LazaUK
MCP Client and Server apps to demo integration of Azure OpenAI-based AI agent with a Data Warehouse, exposed through GraphQL in Microsoft Fabric.
Overview
What is MCP Connector: Integrating AI agent with Data Warehouse in Microsoft Fabric?
This is a demonstration repository that integrates an Azure OpenAI‑powered AI agent with a Microsoft Fabric data warehouse using the Model Context Protocol (MCP) and GraphQL. It enables dynamic discovery of tools and data resources for bidirectional access to enterprise data.
How to use MCP Connector: Integrating AI agent with Data Warehouse in Microsoft Fabric?
Configure a Microsoft Fabric data warehouse with a sample table and create a GraphQL API endpoint. Then set up a local environment with the required Python packages (pip install -r requirements.txt) and environment variables: AOAI_API_BASE, AOAI_API_VERSION, AOAI_DEPLOYMENT, and AZURE_FABRIC_GRAPHQL_ENDPOINT. Finally, launch the MCP client by running python MCP_Client_Gradio.py and click “Initialise System” to connect the AI agent.
Key features of MCP Connector: Integrating AI agent with Data Warehouse in Microsoft Fabric
- Integrates Azure OpenAI agent with Microsoft Fabric
- Uses MCP for dynamic tool discovery
- Employs GraphQL as universal data abstraction
