UFO
Microsoft open-source agent framework that automates Windows apps through UI Automation, Win32 and COM plus GUI actions (UFO²), and orchestrates tasks across Windows, Linux and Android devices as dependency graphs (UFO³ Galaxy).
About UFO
Automating multi-step work in Windows desktop apps, or across several machines, needs an agent that can drive native UI and APIs and coordinate dependent tasks.
Who it’s for
- Developers and researchers building Windows desktop automation with LLM agents
- Teams experimenting with cross-device agent workflows across Windows, Linux and Android
When to consider it
Consider UFO when you want an open-source Windows desktop agent that mixes native API calls with GUI clicks, with the option to scale to multi-device orchestration.
Tradeoffs & limitations
- MIT-licensed research project: you supply your own model provider (OpenAI or Azure OpenAI in the quick start, with other models documented) and run it on Python 3.10 or 3.11.
- UFO² targets Windows only; multi-device Galaxy setup is described as moderately difficult and requires running server and client agents on each device.
SOTA overview · Documentation-based assessment · Sources & review method
Updates
No updates shared yet.
Discussion
Newest firstAsk a question or share how you use UFO.
Keep it helpful. Community rules
Loading discussion…
Sources & review method
Documentation-based assessment · Nov 1, 2026 · Prepared with AI assistance; not a hands-on benchmark.
Checked by SOTA · AI-assisted documentation review. Selection advice is our assessment; verify current requirements for your deployment.
Import history & original evidence
UFO on GitHub
Microsoft open-source agent framework that automates Windows apps through UI Automation, Win32 and COM plus GUI actions (UFO²), and orchestrates tasks across Windows, Linux and Android devices as dependency graphs (UFO³ Galaxy).
Based on official pages and announcements checked on 2026-11-01. No hands-on test, performance benchmark or popularity ranking is claimed.