Abdullah Formuli
@abdullahformuli
خبر إطلاق مهارةفهم الفيديو
ملخص المنشور · الإنجليزية
Watch Skill gives coding agents a video input: it extracts frames, reads on-screen text, transcribes speech and indexes the results for follow-up questions. The author shares the open-source package and its iterative THE LOOP workflow.