google-gemini/starter-applets ? reverse-engineered prompt
Reverse engineered prompt
Build me a set of small demo apps that show what Gemini can do with images, videos, and maps.
I want one app for spatial understanding, where I can upload an image and see Gemini identify objects, explain what is in the scene, and draw bounding boxes around things it finds. I want another app for video analysis, where I can load a video and ask questions about what is happening in it, with the app showing useful results in a clear way. I also want a map explorer, where I can search for a place and have Gemini help me explore it using Google Maps.
Keep the interfaces simple and easy to understand, with a clean demo feel. Make them work as standalone apps, and make it obvious where I put my Gemini API key so I can run them locally. Use whatever current Gemini and Google Maps docs you need to make the examples work well.
Are you gonna build this?
make sure you review the code using coderabbit