Microsoft Execution Containers (MXC), now GA

Microsoft Execution Containers (MXC), now GA: https://blogs.windows.com/windowsdeveloper/2026/10/07/microsoft-execution-containers-policy-driven-containment-for-ai-agents/ Devs declare which files and network hosts an agent may use; the container enforces it from outside. An agent cannot be its own security authority: that seems right. I have not tried it, so I cannot judge how fine-grained the policies are. How do you limit what your agent can touch?