Provides a standardized interface for applications to supply contextual information to Large Language Models.
Implements the Model Context Protocol (MCP), an open standard by Anthropic, enabling applications to provide structured context to Large Language Models (LLMs). This reference implementation demonstrates how to deploy an MCP server, either locally via devtunnels or to Azure Container Apps, and integrate it seamlessly with AI agents in Microsoft Copilot Studio and Visual Studio Code's GitHub Copilot. It leverages existing connector infrastructure to ensure enterprise-grade security features like Virtual Network integration and Data Loss Prevention while supporting real-time data access for AI-powered agents.