NVIDIA has released OpenShell 0.1.0, an open-source runtime designed to enforce which systems and data an AI agent can access without rewriting the agent. The runtime combines sandboxed execution, controlled service access, credential management, and formal policy analysis to restrict API operations and protect credentials outside the agent workload.
Organizations such as Cadence, Slack, and Gecko Robotics are adopting OpenShell for various applications, including chip design, enterprise automation, and physical robotics governance. OpenShell includes components like the Gateway, Supervisor, and Sandbox to manage agent fleets, inspect outbound requests against policy, and apply kernel-level filesystem and process controls.
The policy prover uses formal logic to verify that modeled permissions remain within defined boundaries and identify actions that cross them. Teams can build locally with a sandbox and deploy into shared infrastructure using workspaces, compute drivers for Docker and Kubernetes, and trusted middleware for identity services.
OpenShell is part of the broader NVIDIA Open Agent Safety Platform, which extends protection across the application, runtime, and infrastructure layers. The platform supports frameworks such as Codex, Claude Code, Pi, and Hermes, and is designed to work across enterprise applications, frontier research, and physical AI systems.
Source: NVIDIA