Microsoft Agent 365
Microsoft
The control plane for enterprise AI agents
Microsoft Agent 365 is a management layer for the AI agents an organisation runs, built around observing, governing and securing them rather than building them. A central registry in the Microsoft 365 admin center gives admins a single view of agent adoption, activity and health, with lifecycle management, access control and compliance handled through Microsoft Entra and Microsoft Purview. Registry sync with AWS Bedrock and Google Cloud lets IT discover and govern agents running outside Microsoft's own stack. Generally available for commercial customers since May 2026.
Primary Functions
agent
governance
security
productivity
Features
- Central registry giving a unified view of agent adoption, activity and health
- Lifecycle management, access control and compliance via Microsoft Entra and Purview
- Registry sync with AWS Bedrock and Google Cloud to govern agents outside Microsoft's stack
- Real-time visibility into the agentic environment with early risk and performance signals
- Start, stop and delete agents across connected platforms from one place
Limitations
- A governance and control plane, not a tool for building agents
- Enterprise-only - commercial segment, per-user licensing
- Deepest integration is with the Microsoft stack; other clouds get basic lifecycle governance
- Runtime blocking and alerting arrived via Intune and Defender in public preview, later than GA