Automated Windows GUI testing and visual analysis toolkit. From WSL or any Unix shell, drive native Windows applications via Python (pywinauto + OpenCV) — take screenshots, inspect controls, click ...
基于视觉语言模型(VLM)的桌面自动化 Agent:根据用户任务与当前屏幕截图,自动执行鼠标、键盘等操作 ...