AI Agent Sandboxes: Why a Single Container Isn't Enough
How to isolate code written by AI agents: comparing containers, gVisor, and KVM, examining the threat model, and running virtual sandboxes in seconds.
Architecture, networking, and operations notes for microsrv VMs, containers, and scale-to-zero.
How to isolate code written by AI agents: comparing containers, gVisor, and KVM, examining the threat model, and running virtual sandboxes in seconds.
The risks of exposing port 22, the differences between a bastion or jump host and an SSH gateway, and a stable ssh <vm>@msrv.space address across live migrations.