OpenShell is the safe, private runtime for autonomous AI agents.

★ 21,648⑂ 2,115👁 75⑃ 3,038● 2,044▤ 105
RustApache-2.0Created 2026-02-24docs.nvidia.com/openshell/latest/
Download: OpenShell v0.1.8OpenShell-0.1.8 ↓0.1.82026-11-10
+23
Daily
+612
Weekly
+5,690
Monthly
21,648
stars
2,115
forks
75
watchers
3,038
prs
2,044
issues
105
releases

README

[!NOTE]
 🚧 OpenShell is now part of the CNCF! We’re in the process of moving development to ai-openshell and governance to the CNCF. Until the move is complete, please continue to use the NVIDIA/OpenShell repo for code, issues, and pull requests.



 [!IMPORTANT]
 New in OpenShell 0.1.x: a stable release cadence, new isolation primitives, an expanded extension surface, and new APIs. Read the 0.1.0 upgrade guide.

OpenShell is the safe, private runtime for fleets of autonomous AI agents. Agents are most useful when they can read files, install packages, call APIs, and use credentials. OpenShell gives them that capability without giving them unrestricted access to your data, secrets, or network. You declare what each agent can touch in a policy, and OpenShell enforces it.

 How It Works

OpenShell governs what agents can do in two ways: it instruments the kernel to enforce policy on every file access, system call, and network connection at runtime, and it uses formal verification to check what a policy change would allow before it is applied.

- Kernel-level enforcement. Each agent runs in an isolated sandbox. Kernel controls confine which files it can access and which system calls it can make, and every network connection passes through a policy check before it leaves the sandbox. Agents never see real credentials; OpenShell adds them only to requests bound for approved endpoints.
- Formally verified policy changes. Before a policy change is approved, OpenShell uses formal verification to flag risky new access it would grant, such as reaching a new host with credentials or calling a new API method, so those changes wait for human review.

See Architecture for how the gateway, supervisor, and sandbox fit together.

 Quickstart

You need Linux, macOS on Apple Silicon, or Windows with WSL 2 (experimental), plus Docker, Podman, or host virtualization. See the Support Matrix for details.

shell
curl -LsSf https://raw.githubusercontent.com/NVIDIA/OpenShell/main/install.sh | sh
openshell sandbox create --name demo


The installer sets up the CLI and a local gateway. The default sandbox image is minimal Ubuntu with no agent installed. To run a real agent, follow Run Your First Agent: it runs OpenCode against a free OpenRouter model and shows how to approve new access as the agent needs it.

For native Windows support through Microsoft MXC, see the Windows build and runtime guide for setup and current limitations.

 Explore Further

- Sandboxes: images, runtimes, GPUs, and lifecycle.
- Policies: filesystem, network, and process rules, with the advisor and prover for reviewing changes.
- Providers: credentials that work only at approved endpoints, including inference.
- Gateways: the control plane for sandboxes, policy, and access.
- Kubernetes: deploy the gateway with Helm. Your CNI must enforce NetworkPolicy.
- Extensibility: middleware, interceptors, and compute drivers.
- Tutorials: step-by-step policy and agent walkthroughs.
- Prerelease and developmen

Top issues

Discussions

Latest release

OpenShell v0.1.8 v0.1.8 2026-11-10
OpenShell v0.1.8

 Quick install

bash
curl -LsSf https://raw.githubusercontent.com/NVIDIA/OpenShell/main/install.sh | OPENSHELL_VERSION=v0.1.8 sh


 Upgrade notes

- Static provider credentials without placement metadata now resolve only in the Authorization header or through explicitly enabled body/WebSocket rewrite policies, so custom headers such as x-api-key, query parameters, and path slots need an explicit placement declaration. Upgrade the gateway before the supervisor, and update existing imported profiles as needed, since editing reference YAML alone does not update the gateway catal

News & coverage (19)

Hacker NewsOpenShell – A safe, private runtime for fleets of autonomous AI agents▲ 508 · 217 comments · 2026-10-22Hacker NewsOpenShell: Nvidia's sandbox for AI agents like OpenClaw▲ 4 · 1 comments · 2026-03-16Hacker NewsNvidia/OpenShell: Safe, private runtime for autonomous AI agents▲ 2 · 0 comments · 2026-10-02Hacker NewsOpenShell – private runtime for autonomous AI agents▲ 2 · 0 comments · 2026-03-17TMCnetBedrock Data Extends Data Control to Agents as OpenShell Moves to the CNCF2026-11-09noteNVIDIA Announces NemoClaw and Dynamo Recipes. How Will This Change Your Work?2026-11-09noteSafely Containing AI Agent Shell Execution with NVIDIA OpenShell2026-10-23MSNOpenShell proves agent permissions with math, then enforces them in silicon2026-10-22SDxCentralNvidia wraps OpenShell around full-stack silicon to stop AI agents going rogue2026-10-20DATAQUESTCan NVIDIA's new platform keep autonomous AI agents under control?2026-10-08
9 moreSearch Engine WatchNvidia launches OpenShell to keep AI agents’ hands off your stuff2026-10-01thenextweb.comNvidia launches agent safety platform backed by over 100 companies2026-09-28quantumzeitgeistRun AI agents longer and safer with NVIDIA’s OpenShell runtime2026-09-28Gadget PilipinasNVIDIA Open Agent Safety Platform Goes Open Source2026-09-29NVIDIA DeveloperAdd Runtime Controls to AI Agents with NVIDIA OpenShell | NVIDIA Technical Blog2026-09-28Kingy AINvidia’s Open Agent Safety Platform Puts a Security Boundary Around AI Agents2026-09-28NVIDIA DeveloperDeploy Self-Evolving Agents for Faster, More Secure Research with a Hermes Agent and NVIDIA NemoClaw2026-06-02The New StackJensen Huang and Bill McDermott bet on OpenShell to secure enterprise AI agents2026-05-12Cisco BlogsDefenseClaw is Live!2026-03-30

Snapshot: 2026-11-16T09:23:14Z · Data license Apache-2.0 · JSON