I wouldn't call your critique negative at all — if anything, it's really constructive. Honestly, all your observations are spot on, and I really value this kind of feedback, especially when it comes to UI/UX, since I'm still taking baby steps in that field 😅.
So, first of all: I'll definitely give the overlay some more love in the next update, and the UI issues you pointed out should be fairly straightforward to tackle.
For the cards in hand, there's actually already an "inspector" mapped to the mouse wheel — it just currently works in that zone only. I'll extend the same behavior to the other zones too, except for the Stack. The Stack will have to wait a little longer, since I'm still figuring out what I want to do with it.
I'll try to get those changes into the hotfix I have planned for this week. If they're as easy to tackle as I expect, the update will land probably Thursday evening/night (CEST).
The other two are a bit trickier, mostly because I'm getting tangled up in the order in which I want to tackle things.
For healing, I want to be a bit cautious. Since you mentioned testing the latest cards, I assume you were playing with the Meissa Deck — if so, I don't think healing is necessarily the solution. I think her deck suffers from a few other design issues, and I've already gone back through some of the Ebony cards and reworked them to see if I can improve the deck and make it more fun to play without messing too much with its identity. But yeah, if that doesn't do the trick, I'll definitely explore other options.
Then, you're absolutely right about the missing explanations, and about the tutorial already lagging behind. They are on the radar, and I'm hiding behind two big reasons here. First, those things require me to "solidify" the rulebook and commit to the wording — and, admittedly, I'm a bit scared to do that right now. That said, in the next update, I'll add a couple of pages—including the controls—to the tutorial as a cheap little patch. And your concern gives me the idea to make the very same tutorial accessible right from the battle, maybe in a compact version, where you can quickly look up indexed keywords. Being mostly plain text should be a compromise that doesn't add too much strain on the maintenance timeline.
The second reason is related to that: I really want to improve the effect descriptions and add an oracle to the card inspector to improve the overall experience and consistency. But—big "but"—both of those are currently gated behind the DSL revision, which in turn is gated behind collecting more examples and edge cases. That mostly means wrapping up the six demo decks, at least.
Technically, you can already see me messing with that process. The fact that "On Anchor" is bold on the cards comes from me parsing the current handwritten text — something I eventually want to have generated directly by the card itself, and therefore something the system will already know about. The dynamic text, when it turns purple, is another early experiment, and one I already think is, more often than not, more confusing — or just plainly wrong (e.g. Cited Reference) — than helpful. So... very early, very wrong explorations. Luckily, I'm pretty sure most of that code won't survive into the 0.2.x versions 😅.
For now, all I can really do is ask for your patience and understanding. The project is still at a very, very early stage, but it is moving forward, and I really do care about making these improvements ASAP.
And seriously, thanks a lot for taking the time to write all of this up. Again, this kind of feedback is incredibly useful ❤️!