I propose the following fixed test cases, possibly shuffling the chars of each one:

import itertools
print(list(map(''.join, itertools.combinations_with_replacement('nswe', r=10))))
# (length 286)

for js/py at least
could hard-code with expected since they don't have ref solutions

forks:

https://www.codewars.com/kumite/670cccade45f1ab15e20cea0?sel=670cccade45f1ab15e20cea0

https://www.codewars.com/kumite/670cce2289e815842c20ce97?sel=670cce2289e815842c20ce97

dropped cause horrible descripcion

It feels like a 4kuy+ kata, in term of time spent using JS

You should write about performance requirement at the very top of your kata

I think the issue is that his JS solution is incorrect, but still passes the tests consistently.

Why is this an issue? What has that test in particular that you think it should be there?

Very genius 😂😂😂

The solution Codewars needs, but not the one it deserves.

Didn'n like it. As well as many new katas 8 and 7 kyu. Confusing description, common solution, no enjoyment. I understand that there is no other way to create new katas, just a cry from the heart.