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

# Use Windsurf with AI Workbench

## Overview

**Using Windsurf with AI Workbench is virtually identical to using VS Code or Cursor except for direct integration.**

Windsurf has the same features for working on remotes and attaching to containers.
However, there is one key difference.

You can't (yet) configure Windsurf as a [Native App](/ai-workbench/reference/glossary#native-app) and then launch it from the **Application Launcher**
Instead, you must attach it from within Windsurf. It takes three steps.

**Windsurf has Remote-SSH and Dev Containers built-in, simplifying setup.**

Unlike VS Code and Cursor, Windsurf ships with these extensions already integrated.
You don't need to install any additional extensions for remote development.

**You need to start the project container first, then attach a Windsurf window to it.**

The lack of direct integration requires three steps.
Start the project container, then attach Windsurf to it by connecting to the remote and then the container.

**To persist the Windsurf server and extensions, install them in the container with a build script.**

This is identical to how it works with VS Code and Cursor. You just need to add the correct code snippet to the `postBuild.bash` script.

## Prerequisites: Windsurf Installation

**Step One: Install Windsurf on the same system you have the Desktop App.**

Go to the [Windsurf site](https://codeium.com/windsurf/download) and download the version for your operating system.

**Remote-SSH and Dev Containers extensions are built-in.**

Unlike VS Code and Cursor, Windsurf includes these extensions by default.
No additional extension installation is required for basic remote development.

<Note title="Windsurf only requires WSL configuration for full local install on Windows.">

**Configure the NVIDIA-Workbench distribution.**

1. Select **Windsurf window > File > Preferences > Settings**
2. Type **Settings Tab > Search Bar >** `dev.containers.executeInWSL`
3. Select **Execute in WSL**
4. Type **Execute in WSL Distro >** `NVIDIA-Workbench`

</Note>

<Note title="Using Podman with Windsurf requires configuring the docker path.">

**Do the following in a Windsurf window.**

1. Select **File > Preferences > Settings**
2. Type **Settings Tab > Search Bar >** `dev.containers.dockerPath`
3. Type **Docker Path >** `podman`

</Note>

## Attach to a Local Container

**Step One: Start the project container.**

1. Select **Location Manager > Location Card > Location Window > Project Card**
2. Select **Project Tab > Dashboard > Project Container > Start Container**

**Step Two: Connect a Windsurf window to the container.**

1. Type **Windsurf window > Search Bar (Top) >** `> Dev Containers: Attach to a Running Container`
2. Hit **Enter**
3. Select **Drop Down > container name**
4. Hit **Enter**

**Step Three: Open the project folder in the Windsurf window.**

1. Select **Windsurf window > Open Folder (LHS)**
2. Locate the project folder and select it

## Attach to a Remote Container

**Step One: Start the project container.**

1. Select **Location Manager > Location Card > Location Window > Project Card**
2. Select **Project Tab > Dashboard > Project Container > Start Container**

**Step Two: Connect a Windsurf window to the remote container.**

1. Type **Windsurf window > Search Bar (Top) >** `> Remote-SSH: Connect to Host`
2. Hit **Enter**
3. Select **Drop Down > location-name or IP**
4. Type **Windsurf window > Search Bar (Top) >** `> Dev Containers: Attach to a Running Container`
5. Hit **Enter**

**Step Three: Open the project folder in the Windsurf window.**

1. Select **Windsurf window > Open Folder (LHS)**
2. Locate the project folder and select it