Transform the way you build with powerful agentic features from Gemini Code Assist’s agent mode, now available to everyone both in VS Code and IntelliJ, with a continuously improved experience across your IDE and CLI.
Gemini Code Assist's Agent Mode helps you tackle complex, multi-step tasks with an intelligent, collaborative approach.
For example, imagine you want to refactor the model, view and controller of a shopping cart to display applied discount codes. This would typically require changing multiple files. With Agent Mode, you simply describe the goal, and Gemini will propose a detailed plan for you to review and approve before any code is changed, giving you complete control over every step.
By combining the power of AI with your expertise, Agent Mode helps you achieve a more efficient workflow and produce better code, while saving you significant time and effort.
Agent Mode was previously launched in the Insiders channel for VS Code. During this time, our team has been working hard to improve the experience and we’re happy to share that Agent Mode is now in Preview and available to everybody.
Agent Mode in VS Code is a reimagined chat experience in Code Assist powered by the Gemini CLI to make developers more productive. It expands the capabilities of simple command/response interactions inside Code Assist chat by providing support for multiple file edits, full project context, built-in tools and integration with ecosystem tools (using MCP) with integrated Human in the Loop (HiTL).
With this launch, several enhancements to Gemini Code Assist Agent Mode in VS Code have been made:
Agent Mode is now also available on Stable release in IntelliJ.
You can take advantage of this by using the Gemini Code Assist chat in Agent Mode to complete complex, multi-step tasks and goals with complete control over every change. Before making any modifications, the agent will present a plan for your review. You can edit, ask for changes, approve, or deny any suggested changes. This collaborative approach combines the power of AI with your expertise, resulting in better code and a more efficient workflow.
You can also choose to enable auto approve mode to let the agent act on your behalf. Once the agent is done you can review and roll back changes as you see fit.
Boost your productivity by a refined Gemini Code Assist experience on your IDE.
You can now have more control over your interactions with chat due to the new ability to stop in progress chat responses, so you can immediately halt undesired long or errant chat responses. And with the newly introduced thinking tokens for IntelliJ, you can get more insights into Gemini's thought process before it delivers a response.
Additionally, write code faster and more efficiently with our latest improvements to code completion for VS Code, boosting the performance of code completion suggestions.
Last but not least, many of the previously launched preview features are now generally available.
This will ensure stability and reliability for your daily workflows.
Since the launch of Gemini CLI’s launch late June, the community has contributed over 2800 pull requests, submitted about 3400 issues and received more than 70.000 stars. The Gemini CLI has been rapidly evolving to continue to bring the power of Gemini directly into your command line. In the last month, there were some important launches and updates focused on security, privacy, and continued refinement.
Gemini CLI GitHub Actions has now launched. It is a no-cost, powerful AI coding teammate for your repository. It acts both as an autonomous agent for critical routine coding tasks, and an on-demand collaborator you can quickly delegate work to. It’s now in beta, widely available.
Triggered by events like new issues or pull requests, it works asynchronously in the background, using the full context of your project to automatically handle tasks. It knows your code, understands what you want to do, and gets it done.
Three key use cases covered today are:
The Gemini CLI has also been updated with a deeper integration into VS Code. It becomes context-aware of the files you have open and can access selected text. This allows the CLI to provide targeted and contextually relevant suggestions, as the tool understands precisely what you're working on at the moment. With native in-editor diffing capabilities, you can see suggested changes side-by-side and edit them before accepting. To start using this, use the Gemini CLI version 0.1.20 or higher and a one-time setup command. The update aims to improve developer workflows by providing more relevant and flexible suggestions directly within the IDE.
In addition to Gemini CLI Github Actions and VSCode deeper integration, we have launched many improvements and fixes with the goal of improving Gemini CLI’s reliability and consistency, along with other launches such as Custom Slash Commands, allowing you to define reusable prompts for streamlining interactions with Gemini CLI.
You are a vital part of shaping the future of the Gemini CLI, which is why we want to be transparent about our plans and share our progress with you every step of the way. For that, we have published a public roadmap for Gemini CLI v1.
We hope these changes make for a better coding experience and make you more productive.
Try Gemini Code Assist for free, and learn more about our recent updates in our release notes.
Gemini CLI + VS Code: Native diffing and context-aware workflows
Unleashing new AI capabilities for popular frameworks in Firebase Studio
Announcing Imagen 4 Fast and the general availability of the Imagen 4 family in the Gemini API
URL context tool for Gemini API now generally available
Meet Jules’ sharpest critic and most valuable ally