Run Untrusted AI Agent Code Safely with Azure Container Apps Sandboxes (opens in new tab)

Microsoft has Microsoft.App/SandboxGroups runs untrusted code generated by agents in hardware-isolated environments. Each sandbox starts from an OCI disk image in less than a second. It can scale to thousands of instances at once and incurs no cost when idle. This billing model suits the short, bursty tasks typical of agentic workloads. The risk is not theoretical. When an LLM generates code and an agent executes it in-process, the execution surface becomes the attack surface. A planner in Py...