The last two sample tests need to be changed :

n_vals = [[1, 0], [2, 3], [3, 18], [4, 60], [5, 150], [342, **420115062**], [56753, **386209565**]]

Cool one!

Another curious number :496, there are so few...

.

(python)

• The sample tests require a function named pareto and the expected output is a list of list

• The submissions tests (also in the description) require a function named pareto_front whereas the expected output is a set of tuples.

• Th domination criteria is not explicit Wiki:

  An allocation {x_{1},\dots ,x_{n}}, where x_{i} in R^{k} for all i, is Pareto-optimal if there is no other feasible allocation
  {x_{1}',... ,x_{n}'} where, for utility function u_{i} for each agent i, u_{i}(x_{i}') >= u_{i}(x_{i}) for all i in {1, ... ,n} with
  u_{i}(x_{i}') > u_{i}(x_{i}) for some i

Certainly not now but I made this suggestion before the kata got approved.

You are import pyramid but the initial solution uses a function named Pyramid

Nice idea, but the returned index should be 0-based

• underspecified task (in particular, the fact that output must be prefixed by #, 0 padding, etc.)
• no sample tests
• no random tests

No random tests : Link

• No random tests
• Should specify y is not considered a vowel
• This kind of task has been done several times, consider adding something original before it gets retired
• You seem to consider _ a punctuation mark
• You mention in the description Action Algorite, maybe you thought Algorithm

• 10 random tests are not enough to prevent false positive solutions.
• Edge cases should be specifically tested.

• n is chosen to be bigger than len(str(isqrt(x))).
• "edge" case s = "12.345", n = 3 added.

Condition about x removed. (Isn't that the same "issue" as the one below?)

Condition about x removed.

Corrected.