I think adding a border is an attempt to preserve the essential parts of the images in those situations. It really should not be necessary, but alas modern/stupid problems require modern/stupid solutions.
I did a similar thing for generating a video from multiple screen captures - GitHub repo here: https://github.com/KaliedaRik/sc-screen-recorder
Could you tell us more about how you made it? What libraries did you use, or did you code it with an AI?
The only thing I think I would need to switch is canvas size, ideally where I can select sizes for Twitter, Instagram etc.
Being able to paste an image would be nice too.
Nice work!
This process is repeated daily by designers, frontend devs, indie builders who creates products, ui snippets that they want world to see and grab their attention. as if you share a normal screenshot, most of the people would simply ignore it. in the sense, sometimes if you present your output wrapped in a good wrapper people would perceive it to be of high value and assess further results as such.
hopefully that clears up!
https://github.com/AlexanderVanhee/Gradia
The border can look nice in certain settings, the annotation tools are handy for drawing attention to specific elements. It has the ability to hide/mask things too. All wrapped in a nice intuitive interface.