HermesClaw (OpenShell)

Hermes AI agent hardware-sandboxed by NVIDIA OpenShell, blocking rogue skills at the kernel level.

Desktopfreeglobal

This community project runs the Hermes AI agent inside NVIDIA OpenShell, which hardware-enforces network, filesystem, and syscall limits at the kernel level so even a fully compromised agent can't escape its sandbox, while preserving Hermes's full memory and skill capabilities. It's aimed at developers running autonomous AI agents who want kernel-level containment beyond application-level permission checks, with switchable security policy presets.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of HermesClaw (OpenShell) - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim HermesClaw (OpenShell) to unlock →

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether HermesClaw (OpenShell) (or something else) actually fits.

Try a problem