There are only 3 cheats currently :
-npc lords are spawning with 25% of their part size is filled (this will decrease to 10% soon)
-npc lords get extra small amount of xp to troops daily (starting with 1.3.0) to allow them have more elite troops otherwise they have 50% tier-1 troops
-npc lords know your garrison sizes at your distant towns
There are no other cheats.
Hey Mexxico, and thank you for your hard work and communication. So far my opinion on most changes you've proposed and implemented have been positive, they have all been good for the game. But I want to provide some criticism regarding npc lords knowing garrison sizes at distance, not because it's "unrealistic", but because I don't believe this "cheat" helps the AI at all. Here's my reasoning:
One of the reasons (out of many reasons), according to my experience, with why factions on the defensive against a stronger faction keeps on losing without the ability to recover (snowballing), is that the AI thinks that besieging a weakened town/castle deep inside enemy territory will shift the balance back in its favour. There are two problems with this assumption:
1) By leaving homeland and venturing far into enemy territory, the faction on the defensive leaves its own lands open for attack by the enemies armies, without being able to be close-by to rush to the defense of its remaining cities and castles.
2) Even if the faction on the defensive successfully manages to locate, besiege and conquer a weakened city or castle, odds are since this new territory might be within the enemies heartlands, it will just soon be retaken by the stronger faction with more resources available to it anyways.
If, instead of knowing the garrison size of towns and castles far away, the NPC's had to work with the information it's able to scout (and remember after scouting for a certain period of time, say a week or two of in-game time? Perhaps more?), it would more usually than not make the decision to attack castles and cities bordering its own territory. This would result in three things:
1) The faction on the defensive would much more quickly be able to respond to the attacking factions attacks against its own territory
2) Army cohesion would on average be higher at the time of besieging an enemy town/castle, leading to sieges not failing due to cohesion running out
3) If the faction on the defensive successfully managed to conquer a bordering city/castle, it would be able to respond much more quickly to rush to its defense once the attacking faction attempts to retake said city or castle.
I think you should take these things into consideration, try them out a little bit and see how they play out when two factions (with one being weaker than the other) are at war with one another over numerous games over some periods of time, and see if perhaps this approach would be more logical and result in more intelligent behaviour by the defending faction.
Edit:
Perhaps a bit unrelated but I think this is important to point out: The autocalculation bonus given to Cavalry of 1.3x over other unit types is one of the major contributing factors to snowballing by Vlandia, Khuzaits and to a certain degree also Battania (which for some reason has alot of cavalry in its party templates).
As far as I know, this bonus to autocalculation is still given to cavalry even during sieges. Which besides not making much sense, makese these cavalry-heavy factions even more "overpowered" during sieges scenarioes, when "realistically" the power of their cavalry should really only shine during field battles, and not during sieges.
In my opinion, the Cavalry bonus should just be removed alltogether, but if this is not on your radar at all, then at the very least consider removing the Cavalry bonus during Sieges, as that would slow down these two powerful factions (Khuzaits and Vlandian) a little bit when it comes to snowballing during Sieges.
Thank you for your hard work and I hope you will take this feedback into consideration.