you should not mutate the tactics dictionary

You could have mentioned compound interest in instructions!!!

The order of letters in the result is not important.

Need clarity, as instructions are not clear on direction or degree to solve:
How does 'gbgbh' => 'ci'.
Steps:

1. gbgbh
2. hbbh
3. hch
4. ic

I got 'ic' not 'ci'. please explain if any mistake in my understanding.

So we have the initial sequence (seq = [0, 1, 2, 2]) and we are adding the next natural number after 2 which is 3. The number of 3s added to the list is determined by the integer at the 3rd index of seq (seq[3] == 2). So now we have [0, 1, 2, 2, 3, 3]. Next is 4 and we add three 4s because seq[4] == 3. Next we add three 5s because seq[5] == 3, and so on.

I found the description ambiguous to understand.

sequence advances with adding next natural number seq[natural number] times so now, 3 appears 2 times and so on

What does adding next natural number times to seq[natural number mean] ?

n = 13 perimeter = 23 || given = 25
n = 11 perimeter = 21 || given = 23
n = 18 perimeter = 27 || given = 29
n = 17 perimeter = 27 || given = 29

please check the output for above values!!!

for int_triang(11) -> n = 11

perimeter = 23, count = 14. sample = [8,8,7]

1> As per your description; a > b + c = ?

2> Note that each triangle [a, b, c] is such that a ≤ b ≤ c and the triangles should be sorted.(given at bottom of description).

for the given sample how does this(2) apply?

Add "Space ray" in tactics for 6th Fixed test. else mention in Kata Description to store tactics for future use.

Done

This should be retired

Description is a mess. The task is very simple and so should the description.
Pretty similar to this one albeit that one doesn't specify the precision.

Rounding direction for the numbers exactly between (ending with 25) should be specified and tested. (These numbers can't always be represented as binary numbers with finite precision, so maybe they shouldn't be tested, but this should be documented anyway.)

No random tests