GH Repository · OthersideAI
self-operating-computer
A framework to enable a multimodal model to operate a computer.
- stars
- 10,298
- 30-day movement
- +21/day
- Related entries
- 4
- Connections
- 1
pythonPythonotherpyautoguiopenaiautomation
A Python framework that lets a multimodal model operate a computer, driving interactions through screen input. It is built around OpenAI models and uses PyAutoGUI to act on the machine.
You want an agent that can see and control a computer interface directly rather than working through APIs.
Use it to
- Automate desktop tasks with a vision model
- Experiment with model-driven GUI control
- Prototype agentic computer-use workflows
- Extend the framework for custom automation
For Developers building AI agents that control computers
- Role
- other
- Language
- Python
- Licence
- MIT
- Forks
- 1,418
- Open issues
- 83
- Last push
- 2025-09-19
- Latest release
- v1.0.1 · 2023-12-06
topicsautomationopenaipyautoguimultimodalagentspython