Android Studio Rabbit 2 Launches with Advanced AI Agent Integration, XR Emulators, and Enterprise Capabilities

Google has officially announced the stable release of Android Studio Rabbit 2, bringing a robust suite of production-ready features designed to modernize and accelerate mobile application development. Unveiled alongside the ongoing live streams at the Android Dev Summit, this latest version introduces groundbreaking capabilities centered around developer flexibility, advanced artificial intelligence integrations, immersive extended reality (XR) testing, and enhanced code review workflows.

The release marks a significant milestone in how developers interact with their integrated development environment (IDE), moving beyond traditional coding assistance into fully integrated, agent-driven engineering workflows. By prioritizing adaptability, Google is directly addressing the diverse needs of individual developers, small teams, and massive enterprise organizations alike.

Bring Your Own Agent (BYOA): Integrate Your Preferred Coding Agents

One of the most notable additions in Android Studio Rabbit 2 is the introduction of Bring Your Own Agent, or BYOA. Developer feedback consistently emphasized the critical importance of flexibility within modern engineering workflows, prompting the creation of a system that combines top-tier Android development tools with a wide array of specialized coding agents.

Rather than locking developers into a single proprietary ecosystem, BYOA allows teams to utilize any Architecture Control Protocol (ACP) compliant agent of their choice. Whether an organization requires the advanced reasoning capabilities of a specific model, the cost efficiency of a particular agent for a large-scale project, or compliance with strict enterprise agreements, the new framework accommodates those needs seamlessly.

Developers can now choose from over 40 distinct coding agents, including the Claude Agent, OpenAI’s Codex, Junie, and Google Antigravity. These agents do not operate in a vacuum; they possess deep integration with essential Android Studio tools such as the Android Emulator, native build systems, and Logcat. Furthermore, they are equipped to leverage specialized Android skills to execute complex, multi-step engineering tasks autonomously. Getting started with the feature is straightforward, requiring users simply to open the agent configuration window, select their preferred model, authenticate, and submit their initial prompt.

Build for Android XR with Emulators

As immersive computing continues to evolve, Android Studio Rabbit 2 brings stability to the Android XR Emulators for XR glasses, audio glasses, and display glasses. These tools enable developers to build, iterate, and test complex Android XR experiences directly from their laptops without requiring physical hardware for every stage of prototyping.

The updated display glasses emulator introduces powerful environmental customization options. Developers can now test real physical environments from anywhere by populating the emulator’s surroundings with custom 360-degree photos or live video feeds from a connected camera. Additionally, Google has incorporated new hardware LED indicators to clearly reflect microphone and camera states, enhancing the fidelity of the simulation.

Improvements to the Pairing Wizard also ensure that virtual phones can be paired with virtual audio and display glasses much faster and more reliably. These immersive tools are readily accessible for developers by launching the XR, audio, or display glasses emulators directly through the built-in Device Manager.

Discover, Install, and Manage Android Skills in Android Studio

To maximize the efficiency of autonomous coding agents, Android Studio Rabbit 2 features a centralized skill manager that makes discovering, installing, and managing specialized workflows easier than ever. This interface grants developers granular control over the procedural guidance and specialized operational workflows available to their chosen agents.

The IDE comes pre-loaded with a comprehensive library of Android skills and Firebase agent skills. These capabilities are designed to tackle heavy lifting tasks, such as migrating legacy XML layouts to Jetpack Compose, upgrading to the Android Gradle Plugin (AGP) 9, implementing responsive adaptive layouts for foldables and tablets, or integrating CameraX and various Firebase services. Developers can access the skill manager through the Agent panel’s menu, by typing an ampersand symbol in the chat box, or by navigating directly through the global settings menu.

Agentic Fixes with LeakCanary

Memory leaks have long been one of the most persistent challenges in mobile application development, often requiring hours of tedious profiling and code analysis to resolve. Android Studio Rabbit 2 directly addresses this pain point by integrating LeakCanary with the IDE’s built-in AI agent.

Through this integration, LeakCanary goes beyond simply detecting memory anomalies. It collaborates with the agent to provide detailed narrative explanations regarding the underlying root causes of specific leaks. More importantly, it actively suggests precise, actionable code modifications to correct the issues. Developers can test and leverage this automated diagnostic workflow directly within the LeakCanary task inside the Android Studio Profiler.

Gemini Enterprise in Android Studio

For engineering teams operating under strict regulatory frameworks, Gemini Enterprise continues to provide enterprise-grade governance, rigorous data privacy controls, and custom internal knowledge base integrations within Android Studio. Large organizations can connect the AI assistant directly to their proprietary design systems, internal SDKs, and private repositories, ensuring that generated suggestions and automated reviews remain highly relevant to their specific infrastructure and architectural standards.

Go from Raw Diffs to Guided Code Reviews with Change Review

As agentic tools enable developers to write, refactor, and ship code faster than ever before, a new engineering bottleneck has naturally emerged: the meticulous review and verification of all generated code changes. To alleviate this friction, Android Studio Rabbit 2 introduces AI Change Review, a comprehensive system designed to streamline code verification regardless of where modifications originate within the workspace.

Change Review initiates the process by identifying recommended fixes, where the agent independently reviews its own recent alterations, pending commits, or committed changes under review. Once these initial recommendations are addressed, developers can launch a guided walkthrough. Instead of forcing reviewers to parse raw diff files in traditional alphabetical order, the guided walkthrough intelligently groups and prioritizes related diff chunks across the entire project into logical, thematic units, significantly accelerating the code review lifecycle.

Developers can begin exploring these features immediately by downloading the latest stable version of Android Studio. Community feedback and bug reports can be submitted through the official developer channels, while further demonstrations and technical deep dives remain available through the ongoing live streams and sessions at the Android Dev Summit.