Description does not make clear when to adjust the case of letters, and when not to.

Description examples should be annotated; without explanation they are pretty much useless.

More example tests would then be not strictly necessary. But one whole test is a bit spare. ATM, the single example test is useless in helping reverse engineer the actual requirements.

Changed

That's actually intentional. Emoji are Characters aswell :)

EDIT: I'm not sure if it is benefitial to state which kind of symbols to expect but I added a note to indicate as much.

BTW, it looks like in the 3rd quick example there's an 🅰️ where there should be an A.

EDIT: ok, so you're including emojis in the tests as well. You should explicitly state the character set that will be tested (ie. unicode).

Input and output should be changed to string.

If the requirements are still not clear, please create a new issue. Thank you.

I've removed the offending part of the requirements.

I've added testing against randomly generated sentences and upped the variation of the random word tests.
For everything else I can think of I'd need to include an implementation of the solution.

You know, it sounded like a neat idea when I wrote it, but probably isn't now that you brought it up.
I don't appreciate how you approached me about it but I must concede that you have probably more experience writing kata's on codewars, so I'll take your advice and remove it.

1-25 random tests is not enough, and proper strings should be generated every time. As of now, the tests can still be "cheated" easily.

If we would want to use this function alongside other case changing functions

Except we don't, and we will never do. Why are you trying to throw in worthless requirements which have nothing to do with the task?

Thank you for your feedback.

I have provided some simple random testing to prevent hardcoded solutions.
Not sure how much use it will be to add further randomization as it will polute the tests unnecessary.

I am open to any actionable suggestions, if you have any on top of your mind.

Thank you for your feedback. I am sorry but I can't agree with you.

If we would want to use this function alongside other case changing functions, such as String.prototype.toUpperCase() and String.prototype.toLowerCase(), we would expect it's return behaviour to be ident.

Do you think it would be easier to understand where this requirement comes from, if it is explicitely explained?

Thank you for your feedback, please let me address these issues. English is not my first language.
Are the new Rule Definitions easier to understand?

No random tests.