Google has announced the addition of Agent Mode for Gemini in its latest Android Studio Narwhal Feature Drop Canary release. This feature allows developers to describe complex development tasks and receive an execution plan from Gemini, which then completes the tasks.
With Agent Mode, developers can tackle various tasks such as building projects, fixing errors, extracting hardcoded strings, and implementing new screens. The agent’s proposed changes can be reviewed, accepted, or rejected, with an option for auto-approval when a developer wants to iterate quickly.
Gemini’s default model has a limited context window, but adding the Gemini API key can significantly expand it. Agent Mode also supports MCP (Mobile Code Platform) to unlock additional capabilities and context.
This feature is now available in the latest Android Studio Narwhal Feature Drop Canary release, with plans for further agentic functionality updates in future releases.
Source: https://sdtimes.com/mobile/agent-mode-for-gemini-added-to-android-studio