MArcomage

I'm not sure how I feel about cards that encourage both players to take the time to build and examine a list of existing cards to see what the odds are and determine how to proceed.

I feel that I should do that now to see which modes grant what cards and the chances of each occurring, all in the name of balancing, but I'm not going to do that for a concept that I'm not even sure is implementable. It seems to me that you might have to hand code the list of possible results (aside from zero-cost which is a sortable feature), which also means that you would have to update this card any time an Uncommon card has its cost altered.

I think it's implementable since card cost is already used in cards as Overpower

Overpower pulls a card from those sharing a keyword, and once it selects one it uses one element of its cost (recruits) to grab an integer and use that value in an addition function. Stairs would have to look at every Uncommon card, read its total cost, perform a calculation, determine whether the card passes the test, remember it if it does, count the total number of cards that passed the test, calculate what chances to assign pulling each, and then select one at random. It would be less intensive on the system to have the list prepared ahead of time, and it's still not an elegant solution. Alternately, cards could be assigned hidden values containing a total of the card's cost, making them searchable like keyword cards, but is it worth adding such data to each card for this?

Indeed, such effect is not supported.

How about if it keeps pulling uncommon cards until one of the correct cost pops up? Sounds inefficient and laborious I know, but it could in theory be implementable, right?

And how for example the volunteer works? It has "'cost'=>"Zero"" in code which create cards array. How is defined 'cost' parameter for non-zero cards? I think it's must be possible to call it...