Security for AI and Agent Augmented Coding on Windows | OD853
Ai agents Windows security Microsoft build 2026 Microsoft execution containers Mxc sdk Agent development Sandboxing De Privileged execution Containerization Agent governance Containment Trust boundaries
This Microsoft Build 2026 session examines security for AI agent development on Windows, focusing on built-in OS-level security, observability, and manageability. It covers the Microsoft Execution Containers SDK (MXC) for agent containment, real examples like sandboxing in GitHub CLI, and governance frameworks including containment, detection, identity, and trust. Targeted at developers building production-ready agents who need to implement safe execution boundaries.
Related videos
Build context-aware agents: From data to decisions | BRK240
Building Agents You Can Trust on Windows | BRK262
What's new in Google Cloud's agent platform
Build work-ready agents with Foundry + Work IQ, govern with Agent 365 | DEM340
Open Source Friday: Governance for AI agents with Imran Siddique
Agent Academy Live
What Docker Sandboxes Get Right About AI Security
Run NanoClaw on MacBook Safely with Docker MicroVM Sandboxes
Architecture for running agents in production
Agent development and AgentOps with BigQuery, ADK, and MCP
Implementing DeepMind innovation: Deep research API
