Announcing Model Context Protocol Support (preview) in Azure AI Foundry Agent Service

Model Context Protocol Support in Azure AI Foundry Agent Service
Azure AI Foundry Agent Service now supports Model Context Protocol (MCP), allowing users to easily integrate services and APIs by connecting them to Foundry agents. MCP is an open standard protocol that defines how services provide functions and context to AI models and agents. By leveraging MCP support in Foundry Agent Service, users can streamline the process of building agents and reduce maintenance time.
Benefits of MCP Support in Foundry Agent Service
-
Easily integrate with services and APIs: Connect internal or external services and APIs to Foundry Agent Service without writing custom functions.
-
Enhance AI agents with enterprise features: Enable enterprise-ready features such as Bring Your Own thread storage in Foundry Agent Service.
Code Samples
# Step 1: Import the needed packages
# Step 2: Create AI Project Client and create an Azure AI Foundry Agent
# Step 3: Create a Thread, Message and Run
# Step 4: Execute the Run and retrieve Message
# Step 5: Clean up
Microsoft's Commitment to MCP
At Microsoft Build 2025, Satya Nadella announced Microsoft's partnership with Anthropic to make Model Context Protocol (MCP) a first-class standard across various platforms. Today's preview support for MCP in Azure AI Foundry Agent Service represents a step towards an "open-by-design" AI ecosystem, allowing MCP servers to be seamlessly integrated with Foundry with zero custom code.
Get Started with Azure AI Foundry
- Create with Azure AI Foundry: Start using Azure AI Foundry to leverage MCP support.
- Download the Azure AI Foundry SDK: Access the SDK to begin implementing MCP integration.
- Read Documentation: Explore documentation to learn more about using MCP in Foundry.
- Take Azure AI Foundry Courses: Enhance your skills with Azure AI Foundry learning resources.
- Join the Conversation: Engage with the community on GitHub and Discord to stay informed and connected.