A few weeks ago, we showcased how we’re advancing agentic AI development with Firebase Studio – our cloud-based AI workspace where you can build and launch the next generation of AI-powered apps from one place. We introduced versatile Agent modes, foundational support for the Model Context Protocol (MCP), and integration with the Gemini CLI.
At I/O Connect India, we’re unveiling new updates that help you combine the power of Gemini with these new features to go from idea to app using some of your favorite frameworks and languages. These updates include:
We’re especially excited to make these announcements in India because it’s home to one of the largest communities of app builders in the world. For example, Pankaj Rai, a Google Developer Expert for Android built an entire mentorship platform for students called Mentor Sarthi with Firebase Studio to foster growth and development. He was even able to use Firebase Studio to bring his GlamMEAi personal stylist app to new audiences on the web without needing to learn how to develop for the web! We hope the following updates help developers across the globe continue to innovate with Firebase Studio.
We’ve updated our workspace templates for Flutter, Angular, React, Next.js, and general Web with new AI capabilities that leverage Gemini to help you build and deploy beautiful, powerful apps. When you create a project with an AI-optimized template, your workspace will default to autonomous Agent mode and a redesigned layout to help you get into flow quickly.
You can get assistance from Gemini to help you plan and execute tasks independently without waiting for step-by-step approval. You can still toggle to the “Ask” and “Agent” modes as suited for your tasks, whether you want to generate entire apps, refine the features, run tests, generate documentation, or add new capabilities.
We've also added airules.md file to each template to provide Gemini with specialized instructions to guide code generation and create documentation for each specific type of application. This enables Gemini to have greater awareness of relevant coding standards, error handling methods, dependencies, and development best practices.
Learn how you can use an AI-optimized template to build a Flutter app in Firebase Studio.
Over the coming weeks, we will be updating all of our templates for your preferred frameworks and languages, such as Go, Node.js, .NET, and more.
Adding Firebase services to your apps help extend your app's capabilities. Starting today, if you want to integrate Firebase backend services into your app, all you need to do is prompt Gemini. If you are using App Prototyping Agent for vibe coding or one of the AI-optimized templates, just tell Gemini what you need your app to do and it will recommend the Firebase services. Save time with fewer clicks while Gemini imports any needed libraries or packages, updates your code to integrate the services you need, and guides you through any additional setup or configuration steps. All at no cost to get started.
Learn how to seamlessly integrate Firebase backend services in this deep dive technical post.
Innovation thrives on experimentation, but sometimes you may need a safety net. You can now fork a workspace to create an identical copy, including preserving the entire agent chat history from the original. This lets you create a reliable backup that you can restore at any point in time, allowing you to build new features and try new things confidently without the fear of making irreversible mistakes. You can also fork workspaces when you need to debug and troubleshoot issues without affecting your main development workspace, and make collaboration smoother by sharing your “work in progress” workspace with others.
We’re continuously refining the AI-assisted development experience to give you greater control and flexibility. Many of you build fully-functional apps in Firebase Studio from just a prompt, but we’ve seen that a lot of you actually start outside of Firebase Studio to craft your prompt. Recently, we launched the new “enhance prompts” feature in Firebase Studio. This feature lets you collaborate and refine your prompts and app ideas with Gemini. We've also significantly increased the project upload size to 100MB, enabling you to bring even larger, more complex existing projects into Firebase Studio.
From our recent launches to these new announcements in Bengaluru, we’re dedicated to making AI-assisted development with Firebase Studio more intuitive, faster, and delightful so anyone, anywhere in the world can bring their most ambitious app ideas to life - whether you have decades of coding experience or none at all.
Check out these new capabilities in Firebase Studio and share what you build on X and LinkedIn with #FirebaseStudio!
