I already knew how to do screenshots in Python, and was tired of relearning this stuff in a different language for each project.
So I made an API that takes a screenshot, and now I don't need to learn it ever again :)
https://api.blockspring.com/users/mapsup/blocks/07b6410853fa...
I used to just run that python script manually on my computer. Now I can just call it from my front-end js when users upload images and return back some text about the image.
https://api.blockspring.com/users/pkpp1233/blocks/5a1b66ef20...
But you can do just about anything. Users have done r statistics from js. Or python sentiment analyses (lots of good libraries there) whenever they get user comments in their rails app.
I was thinking about intercepting the requests it makes and making an open source extension that does the same thing, but I haven't had the time yet.
[1] https://chrome.google.com/webstore/detail/search-by-image-by...