Windows-Use: an AI agent that interacts with Windows at GUI layer
14 comments
·September 9, 2025kh9000
akurilin
I recently tried using Qwen VL or Moondream to see if off-the-shelf they would be able to accurately detect most of the interesting UI elements on the screen, either in the browser or your average desktop app.
It was a somewhat naive attempt, but it didn't look like they performed well without perhaps much additional work. I wonder if there are models that do much better, maybe whatever OpenAI uses internally for operator, but I'm not clear how bulletproof that one is either.
These models weren't trained specifically for UI object detection and grounding, so, it's plausible that if they were trained on just UI long enough, they would actually be quite good. Curious if others have insight into this.
philipbjorge
Important is subjective — In the healthcare space, I’d make the claim that most applications don’t expose themselves correctly (native or web).
CV and direct mouse/kb interactions are the “base” interface, so if you solve this problem, you unlock just about every automation usecase.
(I agree that if you can get good, unambiguous, actionable context from accessibility/automation trees, that’s going to be superior)
freedomben
Agreed. I've noticed ChatGPT when parsing screenshots writes out some Python code to parse it, and at least in the tests I've done (with things like, "what is the RGB value of the bullet points in the list" or similar) it ends up writing and rewriting the script five or so times and then gives up. I haven't tried others so I don't know if their approach is unique or not, but it definitely feels really fragile and slow to me
nikanj
Most Electron software doesn't follow accessibility guidelines and exposes nothing over UIA
null
mtVessel
I feel vaguely vindicated that the agent can't figure out how to use the modern Save as workflow, either, and reverts to the traditional dialog.
philfreo
Cool. Reminds me of using SendKeys() in Visual Basic 6 in the 90s
https://learn.microsoft.com/en-us/dotnet/api/microsoft.visua...
anthk
And BeOS/Haiku with the "Hey" command which does literally the same, but far more than key input. You can interact with widgets too. Under Unix, there's xdotool and friends.
electroly
Looks awesome. I've attempted my own implementation, but I never got it to work particularly well. "Open Notepad and type Hello World" was a triumph for me. I landed on the UIA tree + annotated screenshot combination, too, but mine was too primitive, and I tried to use GPT which isn't as good at image tasks as Gemini as used here. Great job!
yodon
Very cool - does anyone know of an OSX equivalent?
Preferably one that is similarly able to understand and interact with web page elements, in addition to app elements and system elements.
CharlesW
There are MCPs that work with the macOS Accessibility stack, like https://github.com/steipete/macos-automator-mcp, https://github.com/ashwwwin/automation-mcp, https://github.com/mediar-ai/MacosUseSDK, and https://github.com/baryhuang/mcp-remote-macos-use.
For web page elements, you could drive the browser via `do JavaScript` or use a dedicated browser MCP (Chrome DevTools MCP, Playwright MCP).
null
tiahura
LLM’s do a pretty good job of using pywin32 for programs that support COM like office.
Using the UIA tree as the currency for LLMs to reason over always made more sense to me than computer vision, screenshot based approaches. It’s true that not all software exposes itself correctly via UIA, but almost all the important stuff does. VS code is one notable exception (but you can turn on accessibility support in the settings)