Few library and service creators seem to think you can treat what they built as a black box. At the very least, they generally come with some advice on usage patterns to avoid.
Writing performant code unfortunately tends to require at least having a basic working model of how all your dependencies work. Otherwise, you tend to find yourself debugging why you ran out of file handles in production.
Most (useful) libraries are managing state. This means the user of the library should have a pretty good idea of how the state is managed or they will call functions in the wrong order.
Well I think that’s because most open sources licenses remove any liability from the creators and have no support outside of volunteers. For this reason a lot of them, and even on closed source projects to add an extra layer before support, provide basic troubleshooting information.
Writing performant code unfortunately tends to require at least having a basic working model of how all your dependencies work. Otherwise, you tend to find yourself debugging why you ran out of file handles in production.