Implement Agent 365 to observe, govern, and secure Al apps, Copilot & Agents

Last Updated 03 Nov 2026

Course Overview

This advanced course is designed to equip security professionals and architects with the knowledge and skills necessary to implement, secure, and govern Al agents using Microsoft Agent 365. The curriculum focuses on the security architecture, governance frameworks, and operational controls required to manage enterprise-scale Al agent deployments securely. Participants will learn how to leverage the Agent 365 control plane to enforce policies, monitor activities, and mitigate risks associated with autonomous and assisted agents.

Duration - 12 Hours

Level - Advanced

Style - Self paced

Course Type - Project Ready with Labs

Certification - No

Hands on Labs - Yes

Solution Areas - AI Business Solutions (ABS), Security, Cloud and AI Platforms, AI in the flow of human ambition (Frontier), Ubiquitous Innovation (Frontier), Establish a trusted and secure platform for AI (Frontier), Build a unified, governed data and AI estate (Core), Power your business with secure AI integrated into your flow of work (Frontier), Create a competitive edge with AI (Frontier), Modernize your data to drive results and leverage AI (Core)

Featured Product - Microsoft 365 Copilot

Course Modules

Microsoft Agent 365 as the control plane

Introduces Microsoft Agent 365 as the centralized control plane to observe, govern, and secure AI agents across your organization, including agents built on Microsoft and third-party platforms. Covers Agent 365 licensing with Microsoft 365 E7, alignment with Zero Trust principles, and how Microsoft Entra, Microsoft Defender, Microsoft Purview, and Windows 365 for Agents extend Microsoft security to agents.

Observe — real-time control for AI agents

Covers how to discover and inventory Microsoft and non-Microsoft agents by using the agent registry in Agent 365. Explains how Microsoft Entra Agent ID gives agents first-class identities, how to register and manage agents, and how to visualize connections between agents, people, and data while addressing data residency and compliance.

Govern — guardrails for AI agents and users

Explains how to use the Agent 365 overview page to manage the agent lifecycle and onboard agents with IT oversight and security policy. Covers managing agent identities and administrative relationships in Microsoft Entra Agent ID, governing ownerless agents, and auditing agent activity with Microsoft Purview.

Secure — comprehensive AI agent protection

Shows how to extend Conditional Access and identity controls to agents, and protect agent interactions with Microsoft Purview DSPM, Data Security Investigations, and the Microsoft Purview SDK for custom agents. Covers defending agents against threats by using AI security posture management, advanced hunting, and threat detection and response in Microsoft Defender.

Agent data access & sharing risks

Examines how agents access and share content, and the oversharing and data exposure risks this creates. Covers using Shadow AI discovery and prompt injection protection in Global Secure Access, and controlling access to Model Context Protocol (MCP) and tooling servers.

Security Dashboard for AI

Introduces unified, real-time visibility into AI risk across Microsoft Entra, Microsoft Defender, and Microsoft Purview, including an inventory of agents, models, MCP servers, and AI apps. Shows how to prioritize risks, assign recommendations, and take remediation actions in the relevant Microsoft Security product.

Security Copilot & Microsoft Security Store

Covers how Microsoft Security Copilot supports agentic AI and how Security Copilot agents extend SOC capabilities such as triage, investigation, and threat hunting. Explains how to discover and deploy Microsoft and partner-built agents from Microsoft Security Store, and how to apply responsible AI principles across your agent strategy.

Foundry Control Plane - AI lifecycle governance

Shows how to use Microsoft Foundry Control Plane to gain visibility and control over agents, models, and tools across your Foundry enterprise. Covers enforcing guardrail policies, and evaluating, monitoring, and optimizing AI apps and agents through fleet monitoring, continuous evaluation, and red-teaming.

Project-Ready Skills Assessment

Take this assessment to validate your skills gathered from the self-paced online learning completed in this course to mark your completion.

Course Completion Survey

Share your feedback with us regarding your experience!

Frontier Engineer Assessment

Take your Frontier Engineer Assessment to mark your completion

This course is part of these learning journeys


Frontier Transformation Engineer - Titan
226 Hours • Complete Learning Journey

Other courses in this Category

Intermediate

Develop secure AI productivity foundation

Duration - 12 Hours
Course
Intermediate

SAP integration with Microsoft Power Platform

Duration - 4 Hours
Course
Intermediate

Accelerate data insights with Copilot in Power BI

Duration - 4 Hours
Course
Intermediate

Implementing Windows 365

Duration - 4 Hours
Course
Earn a digital badge

Complete this course to earn your official digital badge, a verified recognition of your skills and achievement.