Google's Android Studio now integrates Gemini AI agents across the full app development lifecycle. Key features include a new project agent that generates a complete Android app from a single prompt using Jetpack Compose and Material 3, planning mode for complex feature implementation, MCP (Model Context Protocol) support for connecting external tools like GitHub issue trackers, automated dependency updates with build verification, AI-powered UI testing via Journeys with autonomous bug fixing, and Android Skills — markdown-based instruction files that ground LLMs with up-to-date Android best practices. Skills support workflows like XML-to-Compose migration, AGP upgrades, and Navigation 3 integration. All features are available in the latest Canary build and support multiple AI models including Anthropic and OpenAI in addition to Gemini.