OP solved it, closing

Not a kata issue. Also, your code should return the array of original unmodified string instead of the change-cased one in which you use for comparison with values in dictionary.

Your latest solution did not finish the LCM computation for all pre-processed sum values (check your for loop's condition)

Your code fails for cases like n = 225 because it is divisible by 9, resulting in 25. However, there is no decimal point in 25, so p now becomes [25] (an array of string) due to splitting. Trying to access a non-existent index (1) results in undefined, and trying to access undefined[0] results in the Error.

The user's already solved the kata.

Not an issue.

Not an issue.

You'd have a whole other problem if your list is empty other than reduce here using strictly list.length as the denominator.

There are also non-integers.

Object.values doesn't exist in CW's Node 6 version (it's a ES7 thing).

This is a riddle. The whole point is that you're supposed to figure out what the riddle is by investigating different things.

No, exactly k moves is right. ;-)

if the robot returns to the starting point in fewer than k moves, that point should not count as a valid point.

It means such a sitution:
Some robot after 3 moves back to the starting point, after 6 moves it back to the starting point again. and so on.
This robot should NOT be counted when k=6 or k=9 or k=12...

It's using regular expression magic to do the trick:

First, it looks for any characters that aren't newlines (\n), and duplicate them k times. This effectively does this:

abcdef ==> aaabbbcccdddeeefff //assuming k=3

Then it looks for a whole chunk of characters between newlines, and duplicate this chunk n times, inserting new lines inbetween:

abcdef ==> //assuming n=3
abcdef
abcdef
abcdef