/cdx → Azure → model-context-protocol
Model Context Protocol
Part of the scenario: Innovate with Low Code AI and Agents
Model Context Protocol (MCP) enables makers to connect to existing knowledge servers and data sources directly from Copilot Studio. When connecting to an MCP server, actions and knowledge are automatically added to the agent and updated as the functionality evolves. Building agents is simplified and time spent maintaining the agent is reduced. Runs in a live demo tenant with realistic data — you try the real thing, not a slide deck.
More demos in this story
6 relatedAdding Telemetry to your Copilot Studio Agents
This video explores how to add telemetry to your Copilot Studio agents. Learn how to use Application Insights, an Azure service, to monitor your applications with dashboards, search tools, and alerts. It demonstrates connecting Copilot Studio to Application Insights, creating custom telemetry events, and using Kusto queries for advanced log analysis. Perfect for those looking to gain deeper insights into their agents' performance and troubleshoot issues effectively.
Agent feed for users and makers
Agent feed for users and makers demo video
Agent flows
Demo - how to use agent flows in Copilot Studio for deterministic business processes
Agent inventory
This demo video shows how to navigate Copilot Studio agent inventory in the Power Platform admin center.
Agent Store
This demo video provides an overview of the new agent store experience.
Alternative demo scenarios
Extra use cases for demoing how to build agents for M365 Copilot with Copilot Studio