UFO

Microsoft Research's open-source agent stack that operates Windows apps and orchestrates devices.

Open Source Automation & Agents
Visit Website ↗ by Microsoft

What is UFO?

UFO is Microsoft Research's open-source project for UI-focused agents, now spanning three layers in one MIT-licensed repo. The original dual-agent system drives Windows desktop apps through a mix of visual perception and native UI Automation; UFO2 adds a deep-OS Desktop AgentOS runtime; and UFO3 Galaxy orchestrates tasks across multiple devices. Researchers and builders can run each layer independently or compose them into a multi-device automation fabric.

Key features

  • Dual-agent desktop automation (HostAgent + AppAgent)
  • UFO2 desktop agent runtime with deep OS integration
  • UFO3 Galaxy multi-device orchestration over a WebSocket protocol
  • Hybrid vision + Windows UIA perception
  • Published research papers and benchmarks

Pros

  • Research-backed and MIT-licensed
  • Desktop, runtime, and multi-device orchestration in one repo
  • Hybrid GUI+API actions are more reliable than vision-only

Cons

  • Windows-only for the device agent
  • Research project, not a polished product
  • Requires BYO model keys and Python setup

UFO at a glance

PricingOpen Source
CategoryAutomation & Agents
Platformswindows
CompanyMicrosoft
Official websitegithub.com
computer-usewindowsdesktop-automationresearchmulti-device

How to choose the right Automation & Agents tool

Automation and agent tools promise to do work for you, so judge them on reliability before cleverness. Map the exact workflow you want to automate, then check the tool's integrations cover every app in that chain — one missing connector breaks the whole flow. Look for clear triggers, sensible limits, and logs you can read when something goes wrong, because something eventually will. For agents that act on your behalf, approval controls matter: you should be able to review messages, purchases or changes before they happen, especially at the start. Pricing usually scales with runs or tasks, so estimate your monthly volume honestly. Start with one low-risk workflow, measure the time it truly saves, and expand from there.

Compare all Automation & Agents tools →

UFO FAQs

What is UFO?

UFO is Microsoft Research's open-source project for UI-focused agents, now spanning three layers in one MIT-licensed repo. The original dual-agent system drives Windows desktop apps through a mix of visual perception and native UI Automation; UFO2 adds a deep-OS Desktop AgentOS runtime; and UFO3 Galaxy orchestrates tasks across multiple devices. Researchers and builders can run each layer independently or compose them into a multi-device automation fabric.

What are the main features of UFO?

UFO offers Dual-agent desktop automation (HostAgent + AppAgent), UFO2 desktop agent runtime with deep OS integration, UFO3 Galaxy multi-device orchestration over a WebSocket protocol, Hybrid vision + Windows UIA perception and Published research papers and benchmarks.

How much does UFO cost?

UFO is open source, so the software itself is free to download, use and self-host. Your only costs are the infrastructure you run it on and any paid services you connect to it.

What are the pros and cons of UFO?

Strengths of UFO include Research-backed and MIT-licensed; Desktop, runtime, and multi-device orchestration in one repo; Hybrid GUI+API actions are more reliable than vision-only. On the downside, Windows-only for the device agent; Research project, not a polished product; Requires BYO model keys and Python setup.