I really don't know why (googled with no effect), but in some test cases (1-5 in 100 random tests) this solution didn't work. Also, sometimes it gave inconsistent results while running same tests as a "Sample Test" (answer was correct) and in "Attempt" basic tests (incorrect answer in that case). I'm new to programming, but my guess is that it somehow linked to interpreted language nature of Python.
Maybe someone can explain it to me?

Should I kill it?

Thanks for the help, approved!

Good point, added

Yeah, kinda noticed the frequency of the "prime" katas, just wanted to try submitting katas myself for the first time and practice a little in that
That's a clever way to solve it! Thanks!

Got it, thanks!