Fork with new description which may be better

You're welcome, your question is very useful , you opened my eyes to some understatement of the description, I'll try to fix it

Thank you very much for your criticism, I will try to correct the condition. Explaining in more detail to kata, there is a sequence of natural numbers starting with some unknown number . (the numbers go in a row) you need to find the smallest number that will fit this condition , it is guaranteed to exist

That's because your function isn't returning a list.

returning a print statement is the same as returning None. so when the site is testing if your output is correct, it is just testing against a None value.

No problem :)

You are right, there is a problem. I've just published a new fork, I hope it's approved soon.

We don't know what you are doing.

Maybe the random test are wrong or only I

If you don't know, why raise an issue?

All the random tests in Python (I assume you are using this language) just do this :

            test.assert_equals(nines(n), reference_solution(n), f"With n = {n}")

Just like fixed tests do stuff like this

        test.assert_equals(nines(3950), 1035, "With n = 3950")

Please be more specific, and however don't raise issues if you don't know.

prod is new in Python 3.8, here you have 3.6 (for now).