Skip to content

Google introduces updated functionalities for Gemini Code Assist: Universal access granted for Agent mode

Google unveils enhancements for Gemini Code Assist, highlighting the widespread accessibility of the agent mode across Visual Studio Code (VS Code) and IntelliJ for all users. Additionally, there are advancements in IDE integration and developments to the Gemini Command Line Interface (CLI).

Google introduces new functionalities in Gemini Code Assist: Universal access granted for Agent...
Google introduces new functionalities in Gemini Code Assist: Universal access granted for Agent mode

Google introduces updated functionalities for Gemini Code Assist: Universal access granted for Agent mode

Google's Gemini Code Assist development team has made significant strides in improving the functionality of their AI-powered coding tool. As part of Google's broader AI development efforts, which include the release of the Agent Mode feature and the integration of Gemini into multiple IDEs such as IntelliJ and Visual Studio Code (VS Code), these improvements are set to revolutionise the coding experience for developers.

The Power of Agent Mode

One of the key advancements is the Agent Mode, a feature that assists developers with multi-step and complex programming tasks. This mode is now available for all users in both VS Code and IntelliJ, providing a powerful AI coding partner for repositories.

In VS Code, the Agent Mode supports features like inline diff, improved clarity in code changes, persistent history, optimised quota updates, real-time shell command output, and faster UI performance. Meanwhile, the Agent Mode in IntelliJ allows developers to execute complex, multi-step tasks with full control over every change.

Gemini CLI: Your AI Coding Teammate

The Gemini CLI, currently in beta and widely available, is another free and powerful AI coding teammate. It works asynchronously in the background, completing tasks based on the entire context of the project. The updated Gemini CLI in VS Code is better integrated, recognising the context of opened files and accessing selected text for targeted, context-aware suggestions.

Native in-editor diffing functions are now available in VS Code with the updated Gemini CLI, allowing developers to view proposed changes side by side and edit them before accepting.

Other Notable Improvements

Several previously previewed features are now generally available in VS Code, including the ability to return to checkpoints, add selected code snippets to context, invoke Gemini Code Assist with selected terminal output, specify file names in the workspace, and exclude files from Gemini Code Assist use.

Numerous improvements and bug fixes have been introduced to enhance the reliability and consistency of the Gemini CLI. Developers can now formulate target goals, and Gemini Code Assist creates detailed implementation suggestions. Custom slash commands have been added to the Gemini CLI, allowing for the optimization of interaction with the CLI.

Stay Informed and Get Started

Developers can try the Gemini Code Assist for free and learn more about the latest updates in the release notes. A public roadmap for Gemini CLI v1 has been published for transparency. The updated Gemini CLI in Visual Studio Code requires version 0.1.20 or higher and a one-time setup command.

Pause running chat responses in your IDE, and there are new thinking tokens for IntelliJ to provide insights into Gemini's thought process. With these improvements, Gemini Code Assist is poised to make coding more efficient and enjoyable for developers everywhere.

Read also: