I'm glad you liked the intro and aesthetic ^_^
Yes, the trouble is that all cards need to be relatively simple so that they can be combined with others, but that can make them sound uninteresting when they are offered. I think my leading idea to address this is to simply disallow the player from combining a card with itself. By removing Block + Block, the player would be forced to take new cards if they want to take advantage of skill binding. Which should force the player into a more fun play pattern all around, and also make it a lot harder to turbo stack armor. Thank you for highlighting this issue!
Would you say the confusion is due to the fact that cantrip is a keyword? Or that it is unclear how the cost gets determined? The way it works is that the cost of cards is averaged together and rounded up. So a card costing 1 + a card costing 0 = 0.5 which rounds up to 1. Does that part make sense? Does it need to be explained in game? I was hoping being able to see the preview would be enough. I think it can be confusing because I have the cantrip keyword which just poofs. I just worry that it wont be very clear that a card is free without it.
What makes you say that Reckless Assault is a flavor fail? I don't think I want anything which deals damage to automatically be an attack, that is very limiting design-wise and makes skill centric decks not viable. What did you feel like makes a card an attack versus a skill?
Yes, I probably need some means of going back during the Magpie Event. Its a little clunky to implement so it was deprioritized. You can also choose a number that you cannot use, which is a higher priority. I'll try to address both issues when I revisit.
Agreed about Seymours, it's basically impossible to correctly predict. I think I will try your lock in EOT suggestion.
I have also noted down the other small items to look for: anchor interactions, scrap pile bug, conclusive values, stretch outdated wording.
Thank you for bringing up your concerns on the balancing/problematic play patterns which can emerge. I think they can be a little hidden in the playtests because people generally try out different things, but if I don't address it and leave a clear optimal way to min/max the early game, then the replayablity of my game suffers horribly. I really appreciate the insight on that front!!