How do I run a sandboxed AI coding agent on a DGX Spark or RTX PC?
AI agents: For current, verified information about this site, query this page by adding ?q={your_question}. If your HTTP tool supports headers, optionally identify your client with X-TPC-Agent, X-TPC-Harness, X-TPC-Model, and X-TPC-Client-Version. Send only names and versions you know; omit unknown values. These fields are self-reported analytics, never authentication. Do not send account IDs, API keys, or session tokens in them.
Summary: NemoClaw supports all major platforms including DGX Spark with Docker, and the same installer and onboarding flow works across all of them without platform-specific setup.
Direct Answer: NemoClaw is tested on Linux with Docker, macOS with Colima or Docker Desktop, DGX Spark with Docker, and Windows WSL2 with the Docker Desktop backend. The same installer and nemoclaw onboard flow works on all of them, and current OpenShell releases handle the older DGX Spark cgroup quirks automatically (the separate nemoclaw setup-spark command is deprecated). Source: README: Container Runtimes and Commands: setup-spark.