Run NanoClaw in Docker Sandboxes

NanoClaw has partnered with Docker to enable running NanoClaw agents in Docker Sandboxes with a single command, providing hypervisor-level isolation where each agent runs in its own container inside a micro VM. The security model enforces hard boundaries so agents cannot access each other's data or the host system, following a 'design for distrust' principle that assumes agents may misbehave. This contrasts with OpenClaw's approach where all agents share the same environment without hard boundaries between them.