fetchsandbox/playground ? reverse-engineered prompt
Reverse engineered prompt
Build me a small Python playground with a handful of tiny web apps that each connect to a different third party service, and make one app where the integration is already working from scratch.
I want each app to be simple and fast to run locally, with a basic API, clear instructions, and a few realistic bugs hidden in the integrations so an agent can investigate what is broken and prove it with a receipt or note. One app should be a clean setup task where the integration is only partially wired and needs to be finished. Another app should be about testing and finding a planted bug in an existing setup.
Keep the apps lightweight, easy to start, and written in a way that makes it obvious what each one is supposed to do. Add a short README for each app, a main project README, and a simple way to save findings into a markdown file so someone can open a pull request with what they discovered. If you need current docs for the third party services, look them up online.
Are you gonna build this?
make sure you review the code using coderabbit