Source checked · Build

NemoClaw

What the publisher says

Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference.

Best for: Technical teams evaluating supported agents inside an OpenShell sandbox with explicit inference and network controls.

Limitations: Assuming the reference stack is a complete security boundary without reviewing its platform, inference, credential, and network configuration.

Public use: 21,790 GitHub stars · 2,941 forks
Activity: Repository updated 14 Jul 2026
Community: 110 open issues · Discussions available
Checked: 14 Jul 2026

Source facts only. Not tested or recommended by Agent Blocks.

Verify sources

Source Stars Forks Issues Discussions Releases Commits