> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/NVIDIA/OpenShell/llms.txt
> Use this file to discover all available pages before exploring further.

# NVIDIA OpenShell

> The safe, private runtime for autonomous AI agents. Run agents in sandboxed environments that protect your data, credentials, and infrastructure.

NVIDIA OpenShell provides sandboxed execution environments for autonomous AI agents. Agents run with exactly the permissions they need and nothing more, governed by declarative YAML policies that prevent unauthorized file access, data exfiltration, and uncontrolled network activity.

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/quickstart">
    Install the CLI and create your first sandboxed agent in two commands.
  </Card>

  <Card title="How It Works" icon="diagram-project" href="/concepts/architecture">
    Understand the gateway, sandbox, policy engine, and privacy router.
  </Card>

  <Card title="Sandbox Policies" icon="shield-halved" href="/sandboxes/policies">
    Declare filesystem, network, process, and inference constraints in YAML.
  </Card>

  <Card title="CLI Reference" icon="terminal" href="/reference/cli-sandbox">
    Full reference for every command, subcommand, and flag.
  </Card>
</CardGroup>

## Get running in two commands

<Steps>
  <Step title="Install OpenShell">
    ```bash theme={null}
    curl -LsSf https://raw.githubusercontent.com/NVIDIA/OpenShell/main/install.sh | sh
    ```

    Or install via PyPI with [uv](https://docs.astral.sh/uv/):

    ```bash theme={null}
    uv tool install -U openshell
    ```
  </Step>

  <Step title="Create a sandbox and launch an agent">
    ```bash theme={null}
    openshell sandbox create -- claude
    ```

    A gateway is provisioned automatically on first use. The sandbox starts with minimal outbound access — controlled by policy.
  </Step>

  <Step title="Apply a network policy">
    ```bash theme={null}
    openshell policy set my-sandbox --policy policy.yaml
    ```

    Policies are hot-reloadable. No sandbox restart required.
  </Step>
</Steps>

## Protection layers

OpenShell applies defense in depth across four policy domains:

| Layer          | What it protects                                   | When it applies            |
| -------------- | -------------------------------------------------- | -------------------------- |
| **Filesystem** | Prevents reads/writes outside allowed paths        | Locked at sandbox creation |
| **Network**    | Blocks unauthorized outbound connections           | Hot-reloadable at runtime  |
| **Process**    | Blocks privilege escalation and dangerous syscalls | Locked at sandbox creation |
| **Inference**  | Reroutes model API calls to controlled backends    | Hot-reloadable at runtime  |

## Explore by topic

<CardGroup cols={2}>
  <Card title="Gateways & Sandboxes" icon="server" href="/sandboxes/gateways">
    Deploy gateways locally, on a remote host, or behind a cloud proxy.
  </Card>

  <Card title="Providers" icon="key" href="/sandboxes/providers">
    Manage agent credentials — injected at runtime, never written to disk.
  </Card>

  <Card title="Inference Routing" icon="route" href="/inference/overview">
    Keep inference traffic private by routing to local or self-hosted backends.
  </Card>

  <Card title="Supported Agents" icon="robot" href="/concepts/supported-agents">
    Claude Code, OpenCode, Codex, Copilot, and more work out of the box.
  </Card>

  <Card title="Community Sandboxes" icon="box-open" href="/sandboxes/community-sandboxes">
    Use pre-built sandbox images or bring your own container.
  </Card>

  <Card title="Tutorials" icon="graduation-cap" href="/tutorials/first-network-policy">
    Hands-on walkthroughs from first sandbox to custom policies.
  </Card>
</CardGroup>

<Note>
  OpenShell is alpha software — single-player mode. One developer, one environment, one gateway. Multi-tenant enterprise deployments are on the roadmap. Expect rough edges.
</Note>