AIFoundry MCP Connector
MCP Client and Server apps to demo integration of Azure OpenAI-based AI agent with a Data Warehouse, exposed through GraphQL in Microsoft Fabric.
What is AIFoundry MCP Connector?
AIFoundry MCP Connector is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp client and server apps to demo integration of azure openai-based ai agent with a data warehouse, exposed through graphql in microsoft fabric.
MCP Client and Server apps to demo integration of Azure OpenAI-based AI agent with a Data Warehouse, exposed through GraphQL in Microsoft Fabric.
This server falls under the Cloud Services category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- MCP Client and Server apps to demo integration of Azure Open
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx aifoundry-mcpconnector-fabricgraphqlConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use AIFoundry MCP Connector
AIFoundry MCP Connector demonstrates how to wire an Azure OpenAI-based AI agent to a Microsoft Fabric Data Warehouse exposed through a GraphQL API, using an MCP server as the middleware layer so the agent can query and mutate enterprise data through natural language via a Gradio chat interface.
Prerequisites
- Python 3.9+ installed with pip
- Azure OpenAI resource with a deployment configured (e.g. GPT-4o)
- Microsoft Fabric workspace with a sample warehouse and a GraphQL API endpoint created
- Azure credentials with access to both the OpenAI resource and the Fabric workspace
- Git to clone the repository
Clone the Repository
Clone the AIFoundry MCP Connector repository to your local machine.
git clone https://github.com/LazaUK/AIFoundry-MCPConnector-FabricGraphQL.git
cd AIFoundry-MCPConnector-FabricGraphQLInstall Python Dependencies
Install all required packages from requirements.txt, which includes the Azure OpenAI SDK, MCP client/server libraries, and Gradio for the web UI.
pip install -r requirements.txtSet Up Microsoft Fabric GraphQL Endpoint
In your Microsoft Fabric workspace, create a sample data warehouse via 'New item → Sample warehouse'. Then create a GraphQL API via 'New item → API for GraphQL', configure it to expose the Trip table (dbo.Trip), and copy the generated GraphQL endpoint URL.
Configure Environment Variables
Set the four required environment variables. AOAI_API_BASE is your Azure OpenAI endpoint URL (e.g. https://your-resource.openai.azure.com/), AOAI_DEPLOYMENT is the model deployment name, and AZURE_FABRIC_GRAPHQL_ENDPOINT is the URL copied from the Fabric GraphQL setup.
export AOAI_API_BASE="https://your-resource.openai.azure.com/"
export AOAI_API_VERSION="2024-12-01-preview"
export AOAI_DEPLOYMENT="gpt-4o"
export AZURE_FABRIC_GRAPHQL_ENDPOINT="https://api.fabric.microsoft.com/v1/workspaces/.../graphql"Start the MCP Connector
Launch the Gradio client application, which starts the MCP server internally and opens a browser-based chat interface. Click 'Initialise System' in the UI before sending queries.
python MCP_Client_Gradio.pyQuery the Data Warehouse
In the Gradio UI, click 'Initialise System' to connect the Azure OpenAI agent to the MCP server and Fabric GraphQL endpoint. You can now ask natural-language questions about the warehouse data. The agent will generate and execute GraphQL queries automatically.
AIFoundry MCP Connector Examples
Client configuration
Environment variable configuration for connecting the MCP connector to Azure OpenAI and Microsoft Fabric.
{
"mcpServers": {
"fabric-graphql": {
"command": "python",
"args": ["MCP_Server_FabricGraphQL.py"],
"cwd": "/path/to/AIFoundry-MCPConnector-FabricGraphQL",
"env": {
"AOAI_API_BASE": "https://your-resource.openai.azure.com/",
"AOAI_API_VERSION": "2024-12-01-preview",
"AOAI_DEPLOYMENT": "gpt-4o",
"AZURE_FABRIC_GRAPHQL_ENDPOINT": "https://api.fabric.microsoft.com/v1/workspaces/your-workspace-id/items/your-item-id/graphql"
}
}
}
}Prompts to try
Natural-language queries to send through the Gradio chat interface once the system is initialised.
- "How many trips are recorded in the data warehouse?"
- "Show me the top 10 most expensive trips by fare amount."
- "What is the average trip duration and distance in the dataset?"
- "List all trips that started after 6pm and lasted more than an hour."
- "What is the total revenue across all trips in the warehouse?"Troubleshooting AIFoundry MCP Connector
Azure OpenAI returns 401 or 403 authentication errors
Ensure AOAI_API_BASE ends with a trailing slash and matches exactly the endpoint shown in the Azure portal. Confirm your Azure credentials are active by running 'az account show'. The connector uses DefaultAzureCredential, so 'az login' must have been run in the same session.
Fabric GraphQL endpoint returns 403 Forbidden
Your Azure account must have at least 'Contributor' access on the Fabric workspace. In the Fabric portal, go to Workspace settings → Manage access and add your Azure AD user or service principal with the appropriate role.
Gradio UI opens but 'Initialise System' does nothing or shows an error
All four environment variables must be set before launching MCP_Client_Gradio.py. Check the terminal output for specific error messages. A common issue is AOAI_DEPLOYMENT not matching the exact deployment name in Azure OpenAI Studio.
Frequently Asked Questions about AIFoundry MCP Connector
What is AIFoundry MCP Connector?
AIFoundry MCP Connector is a Model Context Protocol (MCP) server that mcp client and server apps to demo integration of azure openai-based ai agent with a data warehouse, exposed through graphql in microsoft fabric. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install AIFoundry MCP Connector?
Follow the installation instructions on the AIFoundry MCP Connector GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with AIFoundry MCP Connector?
AIFoundry MCP Connector works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is AIFoundry MCP Connector free to use?
Yes, AIFoundry MCP Connector is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
AIFoundry MCP Connector Alternatives — Similar Cloud Services Servers
Looking for alternatives to AIFoundry MCP Connector? Here are other popular cloud services servers you can use with Claude, Cursor, and VS Code.
Open WebUI
★ 138.2kUser-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Anything LLM
★ 60.4kThe all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.
LocalAI
★ 46.4kLocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
Nacos
★ 33.0kan easy-to-use dynamic service discovery, configuration and service management platform for building AI cloud native applications.
Xiaozhi ESP32
★ 26.7k本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.
Gateway
★ 11.8kA blazing fast AI Gateway with integrated guardrails. Route to 1,600+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.
Browse More Cloud Services MCP Servers
Explore all cloud services servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up AIFoundry MCP Connector in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use AIFoundry MCP Connector?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.