Maximum Good Triplet Distance

You are given an integer array A. A triplet of indices (i, j, k) is good when i < j < k and both visibility conditions hold:
For every index p with i < p < j, A[p] < A[i] and A[p] < A[j].
For every index q with j < q < k, A[q] < A[j] and A[q] < A[k].
Return the maximum possible index distance k - i among all good triplets. Return -1 if no good triplet exists.

Function
maximumGoodTripletDistance(A: int[]) → int

Examples
Example 1
A = [5,1,4,2,6]
return = 4
The triplet (0, 2, 4) is good. Value 1 is strictly below both 5 and 4, while value 2 is strictly below both 4 and 6. Its distance is 4 - 0 = 4.
Example 2
A = [3,3,1,3,3]
return = 3
Equal-height endpoints block visibility through one another because every interior value must be strictly smaller. One maximum-span good triplet is (0, 1, 3), with distance 3.

Constraints
3 <= A.length <= 200000
-10^9 <= A[i] <= 10^9

Here's the reduction. The first condition says i and j can see each other: everything between them is strictly smaller than both. The second says j and k can see each other. So you need a middle index j, a left partner i that sees j, and a right partner k that sees j, then maximize k - i. A monotonic stack finds visible pairs in O(n). Pop while the top is smaller than the current value, and each popped or remaining top pairs with the current element. For each j, take the farthest visible index on the left and the farthest on the right. The pitfall is equality. Example 2 shows equal values block visibility, so use strict comparisons and handle duplicates by popping or replacing carefully. Also remember adjacent indices see each other trivially, since nothing lies between them. Return -1 if no j has both sides. StealthCoder is the hedge if the stack invariant slips under pressure.

StealthCoder is the hedge for the one pattern you didn't drill. It runs invisibly during the screen share.