> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemoclaw/latest/user-guide/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemoclaw/latest/user-guide/_mcp/server.

# User Guide

## Docs

- [NVIDIA NemoClaw](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/home.md): NemoClaw is an open-source reference stack for running sandboxed AI agents more safely, with a single command and Markdown docs for AI assistants.
- [Overview of NVIDIA NemoClaw](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/about/overview.md): NemoClaw is an open-source reference stack for running sandboxed AI agents more safely inside OpenShell.
- [NemoClaw Architecture Overview](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/about/how-it-works.md): Learn how NemoClaw combines a host CLI, sandbox plugin, and versioned blueprint to run supported agents in a controlled sandbox.
- [Ecosystem](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/about/ecosystem.md): How the OpenClaw, OpenShell, and NemoClaw projects form one stack, where NemoClaw sits, what it adds beyond the OpenShell community sandbox, and when to use the reference integration versus OpenShell alone.
- [Release Notes](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/about/release-notes.md): Changelog and feature history for NemoClaw releases.
- [Prerequisites](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/get-started/prerequisites.md): Hardware, software, and supported platforms for running NemoClaw.
- [Prepare Windows for NemoClaw](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/get-started/prerequisites/windows-preparation.md): Prepare a Windows machine for NemoClaw before running the Quickstart: enable WSL 2, install Ubuntu, and configure Docker Desktop.
- [NemoClaw Quickstart with OpenClaw](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/get-started/quickstart.md): Install NemoClaw, launch a sandbox, and run your first agent prompt.
- [Quickstart with LangChain Deep Agents Code](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/get-started/quickstart-langchain-deepagents-code.md): Create a NemoClaw sandbox that runs LangChain Deep Agents Code as a terminal harness.
- [NemoClaw Inference Options](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/inference/inference-options.md): Inference providers available during NemoClaw onboarding and how the routed inference model works.
- [Use a Local Inference Server](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/inference/use-local-inference.md): Connect NemoClaw to a local model server such as Ollama, vLLM, TensorRT-LLM, or any OpenAI-compatible endpoint.
- [Tool-Calling Reliability for Local Inference](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/inference/tool-calling-reliability.md): Diagnose local inference setups where tool calls leak as plain text and choose when to use Ollama or vLLM.
- [Model Capability Audit Matrix](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/inference/model-capability-audit.md): Maintained matrix template for auditing NemoClaw model and provider behavior across supported agent surfaces.
- [Switch Inference Models at Runtime](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/inference/switch-inference-providers.md): Change the active inference model for a running sandbox without restarting it.
- [Set Up Task-Specific Sub-Agents](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/inference/set-up-sub-agent.md): Where NemoClaw stores OpenClaw sub-agent model configuration, credentials, and workspace files inside the sandbox.
- [Declarative Multi-Agent Manifest](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/inference/declarative-agents-manifest.md): Bake secondary OpenClaw agents into a NemoClaw sandbox image from a checked-in agents.yaml manifest, including per-agent models and OpenClaw-native sub-agent delegation.
- [Manage Sandbox Lifecycle](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/manage-sandboxes/lifecycle.md): List sandboxes, check health, inspect logs, manage dashboard ports, reconfigure providers, rebuild safely, upgrade sandboxes, and uninstall NemoClaw.
- [Runtime Controls and Sandbox Mutability](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/manage-sandboxes/runtime-controls.md): Consolidated reference for what you can change on a running NemoClaw sandbox and what requires rebuild or re-onboard.
- [Messaging Channels](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/manage-sandboxes/messaging-channels.md): Connect Telegram, Discord, Slack, WeChat, WhatsApp, or Microsoft Teams to your sandboxed OpenClaw or Hermes agent using OpenShell-managed channel messaging.
- [Workspace Files](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/manage-sandboxes/workspace-files.md): What workspace personality and configuration files are, where they live, and how they persist across sandbox restarts.
- [Backup and Restore Workspace Files](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/manage-sandboxes/backup-restore.md): How to back up and restore sandbox workspace and state files before destructive operations.
- [Install OpenClaw Plugins](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/manage-sandboxes/install-openclaw-plugins.md): How to install OpenClaw plugins in a NemoClaw-managed sandbox today.
- [Sandbox Image Hardening](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/manage-sandboxes/sandbox-hardening.md): Security hardening measures applied to the NemoClaw sandbox container image.
- [Approve or Deny Agent Network Requests](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/network-policy/approve-network-requests.md): Review and approve blocked agent network requests in the TUI.
- [Customize the Sandbox Network Policy](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/network-policy/customize-network-policy.md): Add, remove, or modify allowed endpoints in the sandbox policy.
- [Common NemoClaw Integration Policy Examples](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/network-policy/integration-policy-examples.md): Guided examples for adding post-install integration policy access to a NemoClaw sandbox.
- [Deploy NemoClaw to a Remote GPU Instance](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/deployment/deploy-to-remote-gpu.md): Run NemoClaw on a remote GPU instance and understand the legacy Brev compatibility flow.
- [Launch NemoClaw with the Brev Web UI](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/deployment/brev-web-ui.md): Launch a hosted NemoClaw sandbox from the Brev web interface without installing the CLI or using a local GPU.
- [Monitor Sandbox Activity and Debug Issues](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/monitoring/monitor-sandbox-activity.md): Inspect sandbox health, trace agent behavior, and diagnose problems.
- [NemoClaw Security Controls, Risks, and Posture Profiles](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/security/best-practices.md): A risk framework for every configurable security control in NemoClaw: defaults, what you can change, and what happens if you do.
- [Credential Storage](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/security/credential-storage.md): Learn where NemoClaw stores credentials, what protections it applies, and how to inspect or rotate stored secrets.
- [Trusted Computing Base for Gateway Lifecycle Control](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/security/trusted-computing-base.md): Defines the trusted computing base, privilege boundaries, and review invariants for NemoClaw gateway lifecycle and shields operations.
- [OpenShell 0.0.71 Gateway Authentication Review](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/security/openshell-0.0.71-gateway-auth-review.md): Reviews the OpenShell 0.0.71 gateway authentication boundary, compatibility behavior, and supporting validation in NemoClaw.
- [OpenClaw Security Controls Beyond NemoClaw's Scope](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/security/openclaw-controls.md): Documents application-layer security controls that OpenClaw provides independently, where NemoClaw adds no additional protection.
- [Platform Support and Launch Claims](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/reference/platform-support.md): Canonical matrix of supported platforms, inference providers, agents, integrations, and deployment paths for NemoClaw, with explicit status for each row.
- [NemoClaw Enterprise Readiness and Admin Capability Guidance](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/reference/enterprise-readiness.md): A reviewed reference for what NemoClaw supports today, what requires manual or admin handling, what is platform-owned, and what is roadmap-only.
- [Architecture Details](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/reference/architecture.md): Learn how NemoClaw combines a host CLI, sandbox integration layer, and versioned blueprint to run compatible agents in controlled OpenShell sandboxes.
- [NemoClaw CLI Commands Reference](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/reference/commands.md): Full CLI reference for standalone NemoClaw commands and agent-specific in-sandbox commands.
- [Choose Between NemoClaw and OpenShell CLIs](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/reference/cli-selection-guide.md): Choose between the NemoClaw CLI and the OpenShell CLI for common sandbox operations.
- [Host Files and State](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/reference/host-files-and-state.md): Reference for NemoClaw host-side files and directories under ~/.nemoclaw.
- [Network Policies](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/reference/network-policies.md): Baseline network policy, filesystem rules, and operator approval flow.
- [Troubleshooting](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/reference/troubleshooting.md): Diagnose and resolve common NemoClaw installation, onboarding, and runtime issues.
- [Use NemoClaw Agent Prompts, Docs MCP Server, and Skills with Your AI Coding Agent](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/resources/agent-skills.md): NemoClaw exposes an MCP docs server and Markdown documentation that AI coding agents can use when they help you install, configure, or operate a sandbox.
- [License](https://docs.nvidia.com/nemoclaw/user-guide/openclaw/resources/license.md): Apache 2.0 license for the NemoClaw project.
- [Overview of NVIDIA NemoClaw](https://docs.nvidia.com/nemoclaw/user-guide/hermes/about/overview.md): NemoClaw is an open-source reference stack for running sandboxed AI agents more safely inside OpenShell.
- [NemoClaw Architecture Overview](https://docs.nvidia.com/nemoclaw/user-guide/hermes/about/how-it-works.md): Learn how NemoClaw combines a host CLI, sandbox plugin, and versioned blueprint to run supported agents in a controlled sandbox.
- [Ecosystem](https://docs.nvidia.com/nemoclaw/user-guide/hermes/about/ecosystem.md): How Hermes, OpenShell, and NemoClaw form one stack, where NemoClaw sits, what it adds beyond a custom OpenShell deployment, and when to use the reference integration versus OpenShell alone.
- [Release Notes](https://docs.nvidia.com/nemoclaw/user-guide/hermes/about/release-notes.md): Changelog and feature history for NemoClaw releases.
- [Prerequisites](https://docs.nvidia.com/nemoclaw/user-guide/hermes/get-started/prerequisites.md): Hardware, software, and supported platforms for running NemoClaw.
- [Prepare Windows for NemoClaw](https://docs.nvidia.com/nemoclaw/user-guide/hermes/get-started/prerequisites/windows-preparation.md): Prepare a Windows machine for NemoClaw before running the Quickstart: enable WSL 2, install Ubuntu, and configure Docker Desktop.
- [NemoClaw Quickstart with Hermes](https://docs.nvidia.com/nemoclaw/user-guide/hermes/get-started/quickstart.md): Install NemoClaw, select the Hermes agent, and launch a sandboxed Hermes dashboard and API endpoint.
- [NemoClaw Inference Options](https://docs.nvidia.com/nemoclaw/user-guide/hermes/inference/inference-options.md): Inference providers available during NemoClaw onboarding and how the routed inference model works.
- [Use a Local Inference Server](https://docs.nvidia.com/nemoclaw/user-guide/hermes/inference/use-local-inference.md): Connect NemoClaw to a local model server such as Ollama, vLLM, TensorRT-LLM, or any OpenAI-compatible endpoint.
- [Model Capability Audit Matrix](https://docs.nvidia.com/nemoclaw/user-guide/hermes/inference/model-capability-audit.md): Maintained matrix template for auditing NemoClaw model and provider behavior across supported agent surfaces.
- [Switch Inference Models at Runtime](https://docs.nvidia.com/nemoclaw/user-guide/hermes/inference/switch-inference-providers.md): Change the active inference model for a running sandbox without restarting it.
- [Manage Sandbox Lifecycle](https://docs.nvidia.com/nemoclaw/user-guide/hermes/manage-sandboxes/lifecycle.md): List sandboxes, check health, inspect logs, manage dashboard ports, reconfigure providers, rebuild safely, upgrade sandboxes, and uninstall NemoClaw.
- [Runtime Controls and Sandbox Mutability](https://docs.nvidia.com/nemoclaw/user-guide/hermes/manage-sandboxes/runtime-controls.md): Consolidated reference for what you can change on a running NemoClaw sandbox and what requires rebuild or re-onboard.
- [Messaging Channels](https://docs.nvidia.com/nemoclaw/user-guide/hermes/manage-sandboxes/messaging-channels.md): Connect Telegram, Discord, Slack, WeChat, WhatsApp, or Microsoft Teams to your sandboxed OpenClaw or Hermes agent using OpenShell-managed channel messaging.
- [Workspace Files](https://docs.nvidia.com/nemoclaw/user-guide/hermes/manage-sandboxes/workspace-files.md): What workspace personality and configuration files are, where they live, and how they persist across sandbox restarts.
- [Backup and Restore Workspace Files](https://docs.nvidia.com/nemoclaw/user-guide/hermes/manage-sandboxes/backup-restore.md): How to back up and restore sandbox workspace and state files before destructive operations.
- [Install Hermes Plugins](https://docs.nvidia.com/nemoclaw/user-guide/hermes/manage-sandboxes/install-plugins-hermes.md): Install Hermes plugins for NemoClaw-managed sandboxes.
- [Approve or Deny Agent Network Requests](https://docs.nvidia.com/nemoclaw/user-guide/hermes/network-policy/approve-network-requests.md): Review and approve blocked agent network requests in the TUI.
- [Customize the Sandbox Network Policy](https://docs.nvidia.com/nemoclaw/user-guide/hermes/network-policy/customize-network-policy.md): Add, remove, or modify allowed endpoints in the sandbox policy.
- [Common NemoClaw Integration Policy Examples](https://docs.nvidia.com/nemoclaw/user-guide/hermes/network-policy/integration-policy-examples.md): Guided examples for adding post-install integration policy access to a NemoClaw sandbox.
- [Monitor Sandbox Activity and Debug Issues](https://docs.nvidia.com/nemoclaw/user-guide/hermes/monitoring/monitor-sandbox-activity.md): Inspect sandbox health, trace agent behavior, and diagnose problems.
- [NemoClaw Security Controls, Risks, and Posture Profiles](https://docs.nvidia.com/nemoclaw/user-guide/hermes/security/best-practices.md): A risk framework for every configurable security control in NemoClaw: defaults, what you can change, and what happens if you do.
- [Credential Storage](https://docs.nvidia.com/nemoclaw/user-guide/hermes/security/credential-storage.md): Learn where NemoClaw stores credentials, what protections it applies, and how to inspect or rotate stored secrets.
- [Trusted Computing Base for Gateway Lifecycle Control](https://docs.nvidia.com/nemoclaw/user-guide/hermes/security/trusted-computing-base.md): Defines the trusted computing base, privilege boundaries, and review invariants for NemoClaw gateway lifecycle and shields operations.
- [OpenShell 0.0.71 Gateway Authentication Review](https://docs.nvidia.com/nemoclaw/user-guide/hermes/security/openshell-0.0.71-gateway-auth-review.md): Reviews the OpenShell 0.0.71 gateway authentication boundary, compatibility behavior, and supporting validation in NemoClaw.
- [NemoClaw Enterprise Readiness and Admin Capability Guidance](https://docs.nvidia.com/nemoclaw/user-guide/hermes/reference/enterprise-readiness.md): A reviewed reference for what NemoClaw supports today, what requires manual or admin handling, what is platform-owned, and what is roadmap-only.
- [Architecture Details](https://docs.nvidia.com/nemoclaw/user-guide/hermes/reference/architecture.md): Learn how NemoClaw combines a host CLI, sandbox integration layer, and versioned blueprint to run compatible agents in controlled OpenShell sandboxes.
- [NemoHermes CLI Commands Reference](https://docs.nvidia.com/nemoclaw/user-guide/hermes/reference/commands.md): Full CLI reference for standalone NemoHermes commands and Hermes-specific in-sandbox commands.
- [Choose Between NemoClaw and OpenShell CLIs](https://docs.nvidia.com/nemoclaw/user-guide/hermes/reference/cli-selection-guide.md): Choose between the NemoClaw CLI and the OpenShell CLI for common sandbox operations.
- [Host Files and State](https://docs.nvidia.com/nemoclaw/user-guide/hermes/reference/host-files-and-state.md): Reference for NemoClaw host-side files and directories under ~/.nemoclaw.
- [Network Policies](https://docs.nvidia.com/nemoclaw/user-guide/hermes/reference/network-policies.md): Baseline network policy, filesystem rules, and operator approval flow.
- [Troubleshooting](https://docs.nvidia.com/nemoclaw/user-guide/hermes/reference/troubleshooting.md): Diagnose and resolve common NemoClaw installation, onboarding, and runtime issues.
- [Use NemoClaw Agent Prompts, Docs MCP Server, and Skills with Your AI Coding Agent](https://docs.nvidia.com/nemoclaw/user-guide/hermes/resources/agent-skills.md): NemoClaw exposes an MCP docs server and Markdown documentation that AI coding agents can use when they help you install, configure, or operate a sandbox.
- [License](https://docs.nvidia.com/nemoclaw/user-guide/hermes/resources/license.md): Apache 2.0 license for the NemoClaw project.