Installing the Cloud Code extension

Before you begin

Before you install Google Cloud Code, confirm that the following tools are installed and set up on your system:

  1. Install and set up Visual Studio Code on your machine.
  2. Install and configure language support.

    The Cloud Code extension works best with the following languages:

Additional dependencies


Ensure the following are installed and added to the PATH environment variable of your machine:

  • Docker client; authenticated with your Docker registry
    • If you're using Windows Home, refer to this guide on how to set up your Docker client.
    • If you're using WSL but cannot install Docker for Desktop, use minikube instead for local development with Cloud Code.

Google Cloud project

If you're using Google Cloud, you'll also need a project.

Projects allow you to collect all of the related resources for a single application in one place as well as manage APIs, enable Google Cloud services, add and remove collaborators, and manage permissions for Google Cloud resources:

Custom versions of dependencies

Cloud Code can install some command-line dependencies for you automatically, and will prompt for your permission to install them, if they're not already installed. If you would prefer to use custom versions of these dependencies, you may opt out of dependency management by setting Auto Dependencies ( to false in Settings (accessible through the Manage menu Manage icon > Settings). Then, install the following on your PATH:

Additional prerequisites for M1 Macs

If you're using Cloud Code with an Apple M1 chip, you'll need to set up Rosetta 2.

If you prefer not to use Rosetta, you can install your own ARM versions of the dependencies as described in Custom versions of dependencies.

Importantly, Google Kubernetes Engine and Cloud Run support x86 images only. While M1 support is being worked on, you can still work with these services when using Cloud Code:

  • Kubernetes: To work with a Kubernetes application when using Cloud Code, you can build your images remotely. The recommended approach is to build your image with Cloud Build, adding a cloudbuild profile to your skaffold.yaml and using cloudbuild as your preferred Skaffold profile.
  • Cloud Run: While you can continue to develop and debug a Cloud Run application locally when using Cloud Code on a M1 Mac, you can't deploy to Cloud Run yet. As a workaround, you can deploy to a prebuilt container to Cloud Run using the Cloud SDK.

Installation steps

  • Install the extension directly from the Visual Studio Code Marketplace.

    Alternatively, select the Extensions view Extension icon in VS Code by clicking on its square icon from the left side taskbar. Search for the Cloud Code extension and click Install.

  • Once the extension has successfully installed, the Cloud Code icons are added to the left side activity bar and ready for use.

Authentication with Google Cloud

If you're developing with Google Cloud, Cloud Code automatically installs Cloud SDK (if not already installed) and uses your active configuration's account by default. You can log in and out of this account by using any Cloud Code view (such as the Kubernetes Explorer view Kubernetes Explorer icon) and choosing the appropriate Cloud SDK action.

Logging in this way takes you through the Cloud SDK authentication process in your web browser. To help with application development with Google Cloud APIs, your Application Default Credentials are also updated during this process.

Link with 'Click here to log into Cloud SDK' under the Google Kubernetes Explorer panel in the Kubernetes Explorer

You can log out of the Cloud SDK using the sign out icon in any Cloud Code view, accessible when you hover over the project name. If you'd like to switch projects in Cloud Code instead, you can select the switch project icon and specify your preferred project within the same account.

Icons to switch projects and sign out of Cloud SDK under the Google Kubernetes Explorer panel in the Kubernetes Explorer

Quick look: What can you do with Cloud Code?

What's next

Getting Support

To send feedback, report issues on GitHub, or ask a question on Stack Overflow.