I fear the issue with a game like this is that logic will be insanely hard to develop, this isn't chess, there's no definite X is better than Y, you're going to have to generalize lots of cards, then, add special exceptions to certain cards, but, for each special exception, you're going to have to make ten more exceptions for other cards when played with this card, and, so on. On top of that, if you're playing in any sort of skilled environment (<=20 ranked, or, the arena), then, you're also going to want to take into account the chance of your draw order, for instance, you may have card X that combos with card type Y, you don't have card type Y, but, knowing your deck contains Z amount of card type Y, there's a J percentage chance you'll draw it, and, you'll have to make a decision based on that.
Over all, seems extremely complicated, don't get me wrong, I've also purchased a lifetime subscription, purely because I love the other two bots I purchased here (Although, stopped using them, due to me switching over to GW2 from WoW (No idea why TheBuddyTeam stopped development for their GW2 bot, I'd totally buy it if it were out now), and, Blizzard removing my Diablo 3 licence (Payment issues)), but, I feel like a fully logical bot is far, far, away, and not something we'll be getting this month, or, hell, this year, if, at all.
EDIT:- That being said, I do agree we'll make leaps and bounds at the start, but, it'll very quickly slow down once we have moderately good logic, which, I don't really think so too hard to acquire (Basically current logic (With a few tweaks, of course, current trading implementing is under-par, doesn't trade obviously positive trades) and a couple hundred card exceptions), at which point it'll be very usable in casual & low-skill (high index) ranked, which, I feel is probably where the developers want the bot to be located at.