Image based code generation

Hi folks, first of all I wanted to say that you’ve really developed an awesome product. I’m heavily promoting it within my peers even though I don’t have any shares :smile:

I was wondering if you’re planning to add image scanning capability to Cursor? I can imagine that it would be gamechanging if for example you could pass a screenshot of button designed in Figma only then to be recreated in code by Cursor.


We will definitely explore vision capabilities and see what’s possible.


Was just going to suggest this feature! My case was a pesky frontend bug that would be much easier to convey via screenshot rather than explaining it to the GPT.

Wonder if there’s any movement on this :point_right::point_left:

You can already attach screenshots to prompts, was using it a little bit, but not extensively

1 Like