Code with Gemini Code Assist

This document describes how you can use Gemini Code Assist, an AI-powered collaborator in Google Cloud, to help you do the following in Cloud Code:

  • Provide guidance to help you solve problems with your code.

  • Generate code for your project.

  • Receive inline suggestions while you're coding.

Learn how and when Gemini for Google Cloud uses your data.

To help you comply with any license requirements for your code, Gemini Code Assist provides source citations when its suggestions directly quote at length from a specific source. To learn more about how and when Gemini cites sources, see How Gemini helps you generate code and cites sources.

Code customization lets you get code suggestions based on your organization's private codebase directly from Gemini Code Assist. Learn how to configure code customization.

This document is intended for developers of all skill levels. It assumes you have working knowledge of VS Code and are familiar with Google Cloud. If you prefer, you can also explore Gemini Code Assist in Cloud Workstations, Cloud Code for IntelliJ, and Cloud Shell Editor.

Before you begin

  1. Before testing Gemini Code Assist capabilities in your code file, make sure your file's coding language is supported. For more information on supported coding languages, see Supported coding languages.

  2. Install the Gemini Code Assist + Cloud Code extension if you haven't already. Cloud Code integrates with Gemini Code Assist in your IDE.

  3. If you prefer to use your IDE behind a proxy, see Network Connections in Visual Studio Code.

Connect to Google Cloud and select a project

In this section, you'll connect to Google Cloud and select a Google Cloud project with the Gemini for Google Cloud API enabled in your IDE.

If you select a Google Cloud project without the Gemini for Google Cloud API enabled, you receive a notification that gives you the option to enable the API from the IDE. Select Enable the API in the notification window to enable the API for your project. For more information, see Set up Gemini Code Assist for a project.

If you prefer to follow the Code with Gemini Code Assist walkthrough directly in your IDE, click Launch VS Code and follow the steps in the walkthrough to connect to Google Cloud and activate Gemini Code Assist.

Launch VS Code

Otherwise, follow these steps:

  1. Launch your IDE.

  2. In the status bar, click Cloud Code - Sign In.

    Cloud Code - Sign In button in the status bar.

  3. When prompted to allow Cloud Code to open the external website, click Open.

  4. Follow the prompts to sign into your Google Account.

  5. When asked if you downloaded Cloud Code from Google, click Sign In.

    You're now connected to Google Cloud.

Next, to select a Google Cloud project that has the Gemini for Google Cloud API enabled, follow these steps:

  1. In the Gemini Code Assist status bar, click Gemini Code Assist.

    The Gemini status bar is available.

  2. In the Gemini Code Assist menu, select Select Gemini Code project.

  3. Select a Google Cloud project that has the Gemini for Google Cloud API enabled.

    Gemini is ready to use.

    The Gemini icon in status bar is set to normal.

Chat with Gemini Code Assist

In this section, you learn how to open the Gemini Code Assist pane and chat with Gemini Code Assist to get an explanation of your existing code.

To begin chatting with Gemini, follow these steps:

  1. Open your code file.

  2. In the activity bar of your IDE, click spark Gemini Code Assist.

  3. In the Gemini Code Assist pane, enter the prompt Explain this code to me and click send Send.

    Gemini uses the code in your code file as a reference to your prompt and responds with an explanation of your code.

    To refer to a specific block of code instead of all the code in the file, you can select the block in your code file and then prompt Gemini.

Reset chat history

Gemini Code Assist uses the chat history for additional context when responding to your prompts.

If your chat history is no longer relevant to what you're trying to achieve, you can reset the chat history: in the Gemini Code Assist pane, click delete Reset Chat.

Generate code with prompts

The following sections show you how to use Gemini Code Assist to perform code transformations like function to create a Cloud Storage bucket using the Quick Pick bar.

Code transformation allows you to use commands or natural language prompts in the Quick Pick bar to request modifications to your code, and provides you with a diff view to show pending changes to your code.

You can also select a part of your code and then prompt Gemini Code Assist for help through the Gemini Code Assist pane, and receive and accept or reject code suggestions while you code.

Prompt Gemini Code Assist with code transformation

  1. In your code file, on a new line, press Control+I (for Windows and Linux) or Command+I (for macOS) to open the Gemini Code Assist Quick Pick bar.

  2. In the menu, using the /generate command, enter /generate function to create a Cloud Storage bucket and then press Enter (for Windows and Linux) or Return (for macOS).

    Gemini Code Assist generates code with the /generate command.

    Gemini Code Assist generates the code based on your prompt in a diff view.

    Gemini Code Assist opens a diff view to show generated code.

  3. Optional: To accept these changes, click Accept.

Optional: Prompt Gemini Code Assist in a code file with a comment

If you prefer, you can also prompt Gemini Code Assist in your code file with a comment by following these steps:

  1. On a new line, enter the comment Function to create a Cloud Storage bucket, and then press Enter (for Windows and Linux) or Return (for macOS).

  2. To generate code, press Control+Enter (for Windows and Linux) or Control+Return (for macOS).

    Next to your prompt text in your code file, Gemini Code Assist generates the code in the form of ghost text.

  3. Optional: To accept the generated code, press Tab.

Optional: Change keyboard shortcut for generating code

If the default keyboard shortcut for generating code isn't working as outlined in the previous section, you can change the keyboard shortcut by following these steps:

  1. In your IDE, click File (for Windows and Linux) or Code (for macOS), and then navigate to Settings > Keyboard Shortcuts.

  2. In the list of keyboard shortcuts, scroll until you find Cloud Code: Generate code.

  3. Click Cloud Code: Generate Code, and then click edit Change Keybinding.

  4. In the dialog that appears, enter your own shortcut.

  5. Press Enter (for Windows and Linux) or Return (for macOS).

    You can now use your newly assigned keyboard shortcut to generate code with Gemini Code Assist.

To learn more about changing shortcuts in your IDE, see Keybindings for Visual Studio Code.

Prompt Gemini Code Assist with selected code using chat

Gemini Code Assist can perform tasks or answer your questions based on the code that you select. To get generated code that's based on a prompt with selected code, follow these steps:

  1. In the activity bar, click spark Gemini Code Assist to open the Gemini Code Assist pane.

  2. In your code file, select a block of code.

  3. In the Gemini Code Assist pane text field, enter a prompt for the selected code.

    For example, select a function in your code and enter the prompt Write a unit test for this function:

    Gemini writes a unit test for a selected function.

    Gemini uses your selected code as reference and responds to your prompt.

Get inline suggestions while you code

While you write code, Gemini Code Assist makes inline code suggestions that you can either accept or ignore. To try this feature, follow these steps:

  1. In your code file, on a new line, start writing a function. For example, if you're in a Python file, write def.

    Gemini suggests code in the form of ghost text.

  2. To accept the code suggestion from Gemini Code Assist, press Tab. Otherwise, to ignore the suggestion, press Esc or continue writing your code.

Optional: Disable inline suggestions

If you prefer to disable inline suggestions in Gemini Code Assist, follow these steps:

  1. In your IDE, select Code (for macOS) or File (for Windows and Linux), and then navigate to Settings > Settings.

  2. On the User tab of the Settings dialog, navigate to Extensions > Cloud Code.

  3. Scroll until you find the Cloudcode: Gemini Code Assist: Inline Suggestions: Enable Auto list, and then select Off.

    This turns off the inline suggestions. You can still press Control+Enter (for Windows and Linux) or Control+Return (for macOS) to manually trigger inline suggestions.

Use code transformation Quick Fix

If there's an error in your code, Gemini Code Assist gives you the option to apply a quick fix to the error with code transformation.

To apply a quick fix in your code file, follow these steps:

  1. In your code file, hold your pointer over the squiggly error line and select Quick Fix, and then select /fix.

    Code transformation quick fix in the IDE.

  2. When the quick fix is applied, a diff view appears. To accept these changes, click Accept.

Prompt with specific files in your workspace with local codebase awareness

Local codebase awareness helps you with high quality code suggestions that are syntactically correct and semantically meaningful within the broader context of your codebase, when you specify files for Gemini Code Assist to use as context.

To specify files in your chat prompt, type @ and select the files you want to specify.

Specify files with local codebase awareness.

To get an explanation on the differences of two files in your codebase, follow these steps:

  1. In the activity bar, click spark Gemini Code Assist.

  2. In the Gemini Code Assist pane, enter the prompt Explain the difference between @YOUR_FILE_NAME_1 and @YOUR_FILE_NAME_2 and press Enter (for Windows and Linux) or Return (for macOS), or Tab. You can also click the name of the file in the list to select the file. Clicking the file name adds the file to your prompt context and opens the file in your IDE.

    Gemini Code Assist responds to your prompt while using the two files you specified for context. Gemini Code Assist also includes the files you specified in Context Sources.

Now that you've specified those files, you can continue asking additional questions or prompts in the same chat history, without having to specify the files again.

For example: In the Gemini Code Assist pane, enter the prompt How can I improve YOUR_FILE_NAME_1? (without the @ symbol) and press Enter (for Windows and Linux) or Return (for macOS).

Gemini Code Assist responds to your enquiry about the file you specified in your prompt.

Optional: Exclude files from context with an .aiexclude file

You can exclude files from being considered by Gemini Code Assist for context by creating an .aiexclude file. An .aiexclude file follows the similar syntax as a .gitignore file, with the following differences:

  • An empty or nonexistent .aiexclude file blocks no files in its directory and all sub-directories.
  • An .aiexclude file doesn't support negation, where you prefix patterns with an exclamation point (!).
  • An .aiexclude file matches greedily on the * character. This will match indiscriminately on directories and files. This .aiexclude file doesn't differentiate between ** or * with regards to folders or files.

Only a single .aiexclude file is supported currently. To configure your .aiexclude file, create a file titled .aiexclude in the root of your workspace folder. If you want to place this file outside of the workspace root, or rename the file, the path may be set in the VS Code settings under "Context Exclusion File". The path value provided in this setting will resolve relative to the root folder of your open workspace.

Examples

The following examples demonstrate how you can configure an .aiexclude file:

  • Block all files named apikeys.txt at or below the workspace directory:

    apikeys.txt
    
  • Block all files with the .key file extension at or below the workspace directory:

    *.key
    
  • Block only the apikeys.txtfile at the same directory as the .aiexclude file, but not any subdirectories:

    /apikeys.txt
    
  • Block all files in the directory my/sensitive/dir and all subdirectories. The path should be relative to the workspace root directory.

    my/sensitive/dir/*
    

Disable code suggestions that match cited sources

Gemini provides citation information when it directly quotes at length from another source, such as existing open source code. For more information, see How and when Gemini cites sources.

To prevent code that matches cited sources from being suggested, you can modify the cloudcode.duetAI.recitation.maxCitedLength setting to 0 in a settings.json file:

"cloudcode.duetAI.recitation.maxCitedLength": 0

Use smart actions

To help you be more productive while minimizing context switching, Gemini Code Assist provides AI-powered smart actions directly in your code editor. When you select your code in your code editor, you can view and select from a list of actions relevant to your context.

To use smart actions in your code, follow these steps:

  1. In your code file, select a block of code.

  2. Next to the selected code block, click lightbulb Show Code Actions.

    Smart actions lightbulb icon appears after selecting a block of code in VS Code.

  3. Select an action such as Generate unit tests.

    Gemini generates a response that's based on the action you selected.

Test other example prompts

After reading through the Generate code with prompts section of this document, try some of the following example prompts.

Get an explanation of code

  1. In your code file, select a function that you want explained.
  2. In the Gemini Code Assist pane, enter the prompt Explain this code to me.

    Gemini uses your selected code as a reference and responds with an explanation of the selected function.

Generate test plans

  1. In your code file, select the code that you want to add unit tests for.
  2. In the Gemini Code Assist pane, enter the prompt Write unit tests for my code.

Get help with debugging code

  1. In your code file, select the code that you want to debug.
  2. In the Gemini Code Assist pane, enter the prompt Help me debug my code.

Make your code more readable

  1. In your code file, select the code that you want to make more readable.
  2. In the Gemini Code Assist pane, enter the prompt Make my code more readable.

    If you prefer to focus on a specific part of your code, then select the preferred part of code before prompting Gemini.

Known issues

The following sections outline the known issues of Gemini Code Assist.

Chat responses may be truncated when they include an updated version of a large open file

To work around this issue, select a smaller section of code and include an additional directive in the chat prompt, such as only output the selected code.

Vim: Cannot accept or dismiss code generation suggestions unless in insert mode

When using the Vim plugin in normal mode, you can't accept or dismiss code suggestions.

To work around this issue, press i to enter insert mode, and then press Tab to accept the suggestion.

Vim: Inconsistent behavior when pressing Esc to dismiss suggestions

When you press Esc, both the IDE and Gemini suggestions are dismissed. This behavior is different from the non-Vim behavior where pressing Esc re-triggers Gemini.

Sign-in attempts keep timing out

If your sign-in attempts keep timing out, try adding the cloudcode.beta.forceOobLogin setting to your settings.json file:

"cloudcode.beta.forceOobLogin": true

License recitation warnings don't persist across sessions

If license recitation warnings don't persist across sessions, refer to the persistent logs:

  1. Click View > Output.

  2. Select Gemini Code Assist - Citations.

Connectivity issues in the Gemini Code Assist output window

If you see a connection error or other connectivity problems in the Gemini Code Assist output window, try the following:

  • Configure your firewall to allow access to oauth2.googleapis.com and cloudaicompanion.googleapis.com.

  • Configure your firewall to allow communication over HTTP/2, which gRPC uses.

You can use the grpc-health-probe tool to test connectivity. A successful check results in the following output:

$ grpc-health-probe -addr cloudaicompanion.googleapis.com:443 -tls error: this server does not implement the grpc health protocol (grpc.health.v1.Health): GRPC target method can't be resolved

An unsuccessful check results in the following output:

timeout: failed to connect service "cloudaicompanion.googleapis.com:443" within 1s

To obtain more details, run the following before grpc-health-probe:

export GRPC_GO_LOG_SEVERITY_LEVEL=info

'Activate Gemini Code Assist' still appears after selecting Gemini Code Assist project

If you selected your project per the instructions in this guide, which includes enabling the Gemini for Google Cloud API, then there may be an issue with the LS server.

To find out more details about your issue, perform the following actions:

  1. In the status bar, click spark Gemini Code Assist and then select Send feedback.

  2. Click Show for any of the log files to see more details and to help troubleshoot the issue.

Leave feedback

To leave feedback of your experience, follow these steps:

  1. In the status bar, click Gemini Code Assist, and then in the Quick Pick menu, select Send feedback.

  2. In the form, fill out the Title and Comments fields.

  3. If you want to share your Skaffold or AI Companion logs, make sure that you select the Send Skaffold logs or Send AI Companion logs option.

  4. Click Submit Feedback.

What's next