Figma, you guys are really cooking with the AI features. It's making workflow really, really special and I've been waiting for this moment for a long time. As I'm using it though, I'm seeing that there are some bottlenecks. I have a couple of ideas. I have a simple idea and a more complex one maybe for the future.
The first one is simply just a prompt history where after you generate an image, you can see somewhere the original prompt that you used to generate that image. Many times when I need to iterate on a generation and I go back to edit the prompt, I can't find the prompt anywhere. I’ve started to copy the prompts that I write out and paste them into a text box every time I generate an image because I don't want to lose the prompt. I feel like it's a simple thing that would really improve the workflow.
The second idea, and this one is a little bit more complex, but I think it's right in line with what you're moving towards, is prompt components with variables. So just like we have design components where you can change out different variables within them, it'd be great to have prompt components that you can build with editable fields as variables, and even variants of the prompt. The prompts could be stored in a library for easy use and organization, and maybe even have a drag and drop feature where you can drag and drop one of your pre-made prompts onto a frame to automatically run the prompt using that frame as a reference image. I imagine this could be something that you guys are already working towards by acquiring the node-based editor Weavy, but just thought I'd throw it out there as an idea.
I'm absolutely loving the new features. I'm a long-time Figma user and I’ve been watching the trajectory closely over the past few years. These new AI developments are something I've been waiting for design softwares to do for many years. You guys are really raising the bar. Thank you