Google Open-Sources ARTEMIS: AI-Powered Automation for Android Devices
ARTEMIS turns prompts into automation:
- 99%+ success rate (!!)
- Integrates w/ Codex, Claude Code, Antigravity
- Automates workflows + captures logs
And the roadmap is VERY interesting:
They want to add an on-device VLM so basically local execution with lightweight edge vision models for privacy-first automation.
You can see this like Codex Computer Use but for mobile.
Possibilities are endless!
It has a MCP integration so basically any coding agent will be able to use it (Cursor, Cordex, Antigravity, Claude Code, etc.)
GitHub repo in the first comment!
🔥 Google has open-sourced ARTEMIS, an AI-powered framework for automating Android devices using natural-language instructions.
You can describe a task, and ARTEMIS can observe the device, interact with UI elements, execute multi-step workflows, and capture execution traces, screenshots, and logs.
The project reports 99%+ task completion on Google’s AndroidWorld benchmark, covering 100+ multi-step tasks. That figure applies to the benchmark—not to every possible Android task.
ARTEMIS also provides MCP integration, allowing AI coding assistants such as Codex, Claude Code, Antigravity, Cursor, Windsurf, and others to interact with Android devices.
And the roadmap gets even more interesting 👀
Google lists lightweight on-device VLMs as planned work, which could enable more local, low-latency and privacy-focused visual automation.
For now, ARTEMIS focuses on Android. iOS support is also listed on the roadmap.
This could become an interesting bridge between AI coding agents and real mobile devices. 🚀