Recently I folded a top pair good kicker hand (pair of Aces) due to the two enemies going all in on the turn, with about 15 big blinds each. One of them had top pair bad kicker but the other had two pair.
Even with the good result, this made me realize I go with "the feeling" and don't know what is the standard way to think about these situations.

I play Spin n Go Flash microstakes

I don’t know if I would consider this a rule of thumb, but finding the critical value of a factor needed for +EV is certainly a good decision criterion. What follows is an example of choosing the critical bet size that can apply to your situation.

If you have aces with a good kicker on the turn, you have 14 outs for a river improvement assuming an unpaired rainbow board and ignoring straights. Two of the outs are for trip aces and 12 outs for a high two-pair, consisting of 3 cards that match your kicker plus 3 cards that math one of the 3 other board cards. Therefore, you have roughly a 14/46 = 30% chance to win assuming villain doesn’t have or hit three of a kind. On the other hand, you may win with just the ace pair.

You can show that the bet size you need for +EV given equity of E, is Pot x E/(1-2E) for E < 50%. For equity of 30%, a bet that is not more than 0.30/(1-0.60) = 0.75 x Pot is +EV.

OP Example: Assume a pot of 20. The maximum turn bet for +EV is 0.75*20 = 15 so you can make an all-in bet with this pot given the effective stack.

Naturally you can't do this analysis while playing and there are other factors –questionable assumptions, number of villains, stack size, position, villain characterization, etc. but the above is thought to be a useful EV analysis whose results are subject to adjustment.