Autonomous agents mark a new inflection point in AI. Systems are no longer limited to generating responses or reasoning through tasks. They can take action: Agents can read files, use tools, write and run code, and execute workflows across enterprise systems, all while expanding their own capabilities.
Application-layer risk grows exponentially when agents continuously improve and evolve. The NVIDIA OpenShell runtime is being built to address this.
Part of NVIDIA Agent Toolkit , OpenShell is an open source, secure-by-design runtime for running autonomous agents such as claws. It works by ensuring each agent runs inside its own sandbox, separating application-layer operations from infrastructure-layer policy enforcement.
This means security policies are out of reach of the agent — they’re applied at the system level. Instead of relying on behavioral prompts, OpenShell enforces constraints on the environment the agent runs in — meaning the agent cannot override policies, or leak credentials or private data, even if compromised.
With OpenShell, enterprises can separate agent behavior, policy definition and policy enforcement. Organizations gain a single, unified policy layer to define and monitor how autonomous systems operate. Coding agents, research assistants and agentic workflows all run under the same runtime policies regardless of host operating system, simplifying compliance and operational oversight.
This is the “browser tab” model applied to agents: Sessions are isolated, resources are controlled and permissions are verified by the runtime before any action takes place.
NVIDIA NemoClaw is an open source reference stack that simplifies installing OpenClaw always-on assistants with the OpenShell runtime and NVIDIA Nemotron models in a single command.
NemoClaw includes an example configuration of OpenShell that defines how the agent should interact with systems. NemoClaw uses open source models like NVIDIA Nemotron alongside OpenShell.
This enables self-evolving claws to run more securely in clouds, on premises or on personal computers, including NVIDIA GeForce RTX PCs and laptops or NVIDIA RTX PRO-powered workstations , as well as NVIDIA DGX Station and NVIDIA DGX Spark AI supercomputers .
Both OpenShell and NemoClaw are in early preview. NVIDIA is building in the open with the community and its partners to enable enterprises to scale self-evolving, long-running autonomous agents safely, confidently and in compliance with global security standards.
Get started with NVIDIA OpenShell and launch a ready‑to‑use environment on NVIDIA Brev , or explore the open source project on GitHub .
Hear from NVIDIA CEO Jensen Huang live on stage at SAP Center. Arrive early to catch the GTC Live 2026 pregame show for an insightful discussion on the latest in AI, accelerated computing and transformative tech with industry leaders.
Key considerations
- Investor positioning can change fast
- Volatility remains possible near catalysts
- Macro rates and liquidity can dominate flows
Reference reading
- https://blogs.nvidia.com/blog/secure-autonomous-ai-agents-openshell/#primary
- https://blogs.nvidia.com/blog/author/aligolshan/
- https://blogs.nvidia.com/blog/secure-autonomous-ai-agents-openshell/#disqus_thread
- China develops new ultra-cold alloy that can reach -273°C without helium — could enable compact cooling for superconducting quantum chips, military equipment, a
- Snap Decisions: How Open Libraries for Accelerated Data Processing Boost A/B Testing for Snapchat
- Into the Omniverse: How Industrial AI and Digital Twins Accelerate Design, Engineering and Manufacturing Across Industries
- California scientists 'FlyTrap attack' on DJI drones demonstrated — patterned umbrellas lure autonomous drones close enough to be captured or even induced to cr
- ABB Robotics Taps NVIDIA Omniverse to Deliver Industrial‑Grade Physical AI at Scale
Informational only. No financial advice. Do your own research.