An AI video editing agent that turns natural-language instructions into planned, tool-driven edits, with reusable Style Skills for consistent storytelling.
Jeomon/Windows-Use
An AI agent that controls Windows at the GUI level, reading the screen through UI Automation and using any LLM to click, type, and run tasks.
Open-source alternative to:Microsoft Power Automate Desktop
UiPath
Automation Anywhere
- Stars
- 470
- Forks
- 0
- Open issues
- 0
- Last commit
- 27 days ago
Key features
- 01Reads the screen through the Windows UI Automation API without a vision model.
- 02Supports 14 LLM providers, including Anthropic, OpenAI, Google, and Ollama.
- 03Controls mouse, keyboard, windows, and virtual desktops.
- 04Runs PowerShell commands and reads their output.
- 05Offers a terminal CLI and sync/async Python APIs.
- 06Includes optional voice input and output via STT and TTS providers.
Use cases
- Automating repetitive desktop tasks in Windows applications that lack an API.
- Prototyping AI agents that operate native Windows software.
- Running GUI workflows inside a disposable VM for testing or research.
Topics
FAQ
- What is Windows-Use?
- An AI agent that controls Windows at the GUI level, reading the screen through UI Automation and using any LLM to click, type, and run tasks.
- Is Windows-Use free to use? What license does it have?
- Windows-Use is open source under the MIT license. Check the license terms before using it in commercial products.
- Is Windows-Use still maintained?
- Actively maintained. The last commit to Windows-Use was on Sep 23, 2026.
- What can I use Windows-Use for?
- Automating repetitive desktop tasks in Windows applications that lack an API. Prototyping AI agents that operate native Windows software. Running GUI workflows inside a disposable VM for testing or research.
- Where can I get Windows-Use?
- The source code and installation instructions are on GitHub: https://github.com/Jeomon/Windows-Use
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
AutoGPT is an open-source platform for building, running, and scheduling AI agents that carry out multi-step workflows from plain-English descriptions or a visual builder.
A toolkit that generates agent-native command-line interfaces for existing software, plus a CLI-Hub registry for installing ready-made ones.
Blender add-on and MCP server that let AI assistants like Claude build, modify and inspect 3D scenes in Blender through prompts.
An open-source autonomous research agent that plans questions, gathers web and local sources, and writes cited reports with any LLM provider.
Lets an AI vision model operate a real iPhone using a $7 ESP32-S3 board that acts as an AirPlay screen receiver and USB HID mouse, with no computer needed at run time.