# NVIDIA OpenShell

## Docs

- [Architecture](https://mintlify.wiki/NVIDIA/OpenShell/concepts/architecture.md): How OpenShell works — the four components, request flow, deployment modes, and the K3s cluster inside Docker.
- [Supported agents](https://mintlify.wiki/NVIDIA/OpenShell/concepts/supported-agents.md): All agents that run in OpenShell sandboxes, how credentials are discovered and injected, and how to launch each agent type.
- [Configure Inference Routing](https://mintlify.wiki/NVIDIA/OpenShell/inference/configure.md): Set up the managed local inference endpoint with provider credentials and model configuration.
- [Inference Routing](https://mintlify.wiki/NVIDIA/OpenShell/inference/overview.md): Understand how OpenShell routes inference traffic through external endpoints and the local privacy router.
- [Introduction](https://mintlify.wiki/NVIDIA/OpenShell/introduction.md): OpenShell is the safe, private runtime for autonomous AI agents — sandboxed execution with declarative YAML policies that protect your data, credentials, and infrastructure.
- [Quickstart](https://mintlify.wiki/NVIDIA/OpenShell/quickstart.md): Install the OpenShell CLI and create your first policy-enforced AI agent sandbox in two commands.
- [openshell gateway](https://mintlify.wiki/NVIDIA/OpenShell/reference/cli-gateway.md): CLI reference for gateway lifecycle commands: start, stop, destroy, add, select, login, and info. Includes openshell status and openshell completions.
- [openshell inference](https://mintlify.wiki/NVIDIA/OpenShell/reference/cli-inference.md): CLI reference for inference commands: set, update, and get. Configure the gateway-scoped LLM inference route used by sandbox workloads.
- [openshell policy](https://mintlify.wiki/NVIDIA/OpenShell/reference/cli-policy.md): CLI reference for policy commands: set, get, list, and delete. Manage sandbox and gateway-global policies.
- [openshell provider](https://mintlify.wiki/NVIDIA/OpenShell/reference/cli-provider.md): CLI reference for provider commands: create, get, list, update, and delete. Providers are named credential bundles injected into sandboxes at creation.
- [openshell sandbox](https://mintlify.wiki/NVIDIA/OpenShell/reference/cli-sandbox.md): CLI reference for sandbox lifecycle commands: create, connect, list, delete, get, upload, download, and ssh-config. Includes openshell forward, openshell logs, and openshell term.
- [Default Policy Reference](https://mintlify.wiki/NVIDIA/OpenShell/reference/default-policy.md): Breakdown of the built-in default policy applied when you create an OpenShell sandbox without a custom policy.
- [Gateway Authentication](https://mintlify.wiki/NVIDIA/OpenShell/reference/gateway-auth.md): Gateway resolution priority chain, authentication modes, connection flow, and credential file layout.
- [Policy Schema Reference](https://mintlify.wiki/NVIDIA/OpenShell/reference/policy-schema.md): Complete field reference for the sandbox policy YAML, including static and dynamic sections.
- [Support Matrix](https://mintlify.wiki/NVIDIA/OpenShell/reference/support-matrix.md): Supported host platforms, software prerequisites, container images, and kernel requirements for running OpenShell.
- [Community sandboxes](https://mintlify.wiki/NVIDIA/OpenShell/sandboxes/community-sandboxes.md): Use pre-built sandboxes from the OpenShell Community catalog, bring your own container image, or contribute your own sandbox.
- [Gateways](https://mintlify.wiki/NVIDIA/OpenShell/sandboxes/gateways.md): Deploy local and remote gateways, register cloud gateways, and manage multiple gateway environments.
- [Manage sandboxes](https://mintlify.wiki/NVIDIA/OpenShell/sandboxes/manage-sandboxes.md): Create sandboxes, connect to them, forward ports, transfer files, and manage the full sandbox lifecycle.
- [Policies](https://mintlify.wiki/NVIDIA/OpenShell/sandboxes/policies.md): Apply, iterate, and hot-reload sandbox network policies on running OpenShell sandboxes without restarting.
- [Providers](https://mintlify.wiki/NVIDIA/OpenShell/sandboxes/providers.md): Create and manage credential providers that inject API keys and tokens into OpenShell sandboxes as environment variables.
- [Write your first sandbox network policy](https://mintlify.wiki/NVIDIA/OpenShell/tutorials/first-network-policy.md): Create a sandbox, observe default-deny networking in action, apply a fine-grained L7 read-only rule, and inspect the audit trail — all in under five minutes.
- [Grant GitHub push access to a sandboxed agent](https://mintlify.wiki/NVIDIA/OpenShell/tutorials/github-sandbox.md): Learn the iterative policy workflow by launching a sandbox, diagnosing a GitHub access denial, and applying a custom policy to grant scoped write access.
- [Run local inference in a sandbox](https://mintlify.wiki/NVIDIA/OpenShell/tutorials/local-inference.md): Route sandbox inference requests to a local model backend using Ollama or LM Studio, with options for GPU-accelerated and cloud-hosted models.