The Secret Vaults of Kas the Betrayer is a puzzly Twine dungeon-crawl with fantasy dwarf lore. I played it to an ending, though it was not a very happy ending, and I suspect that qualifies as an early death rather than the intended finish.
As far as I can tell, The Secret Vaults of Kas the Betrayer is a Twine adaptation of an RPG session or module. It’s set in an underground dwarf kingdom, and you can find various supplies — lantern, “iron rations” — as well as hints of the ancient past. It’s the sort of setting where you can expect a lot of oaken and leathern furnishings. In the portion I saw, there are several combination locks required to get from one area to the next. I got only as far as managing to put one combination lock into a wrong configuration and being killed instead of progressing.
Here are the reasons I then stopped playing rather than restarting the game:
Overall polish wasn’t what I would have liked. Unlike many of the other Twine games this comp, this one doesn’t change the default CSS; that’s not an unforgivable sin, but it goes with some other signs of under-polishing, such as its/it’s errors in the text, various typos, and the persistent misspelling of “thief/thieves” as “theif/theives”.
Genre is not really my favorite. This is on me, but it takes a fair amount to get me to consider standard dwarfs-and-elves-style fantasy at this point, and I need some really first-class world-building hooks to get into it. So I admit a certain amount of bias there.
Difficulty suspending disbelief. I know, I know, it’s a genre game. But one of the first things you encounter is mention of a solid quicksilver mining pick. I am assuming that must be Magicke, because mercury does not freeze at temperatures that humans can usually tolerate for long.
Inconsistent level of agency. There were quite a few times when the game text described my character doing things that I did not direct to have happen (some business with an inkwell and a compass, or choosing to light my surroundings with a glowing dagger). At other times, it required very fiddly input. Sometimes it seemed like my character mostly had a mind of their own, and other times they needed everything spelled out. Either one of those by itself can work, but mixing them together without a clear aesthetic purpose tends to confuse and distance me.
But, more critically than any of those, the combination locks themselves:
1. You can get into a state where you’re engaged in trying to solve them and there are no links that mean “back away and come back to this puzzle later.” It was not obvious that I was going to do any more than look at the puzzle machinery when I clicked the link that got me into that situation.
2. I did not see anything in game, either before or after attempting to interact with either of the two locks I got to, that constituted a hint for them.
3. These facts meant that I spent a long while trying random combinations in the first combination lock before getting frustrated and going to the walkthrough. It turned out that the walkthrough contained the sorts of hints that I might have expected to encounter in the game itself, but not the explicit list of winning actions that is what I usually mean by the word “walkthrough”.
This was all frustrating enough that when I went through it all again a second time, and that second combination lock turned out to kill me for no reason that was obvious from either game or walkthrough, I was ready to be done.
It’s possible that all this made for a pretty fun RPG session, but I think it would take a bit more sanding down to make a really working puzzle game. Specifically, it is not clear that the author had in mind a particular model for how the player was supposed to figure out the combinations. Maybe there are in-game clues that I simply failed to discover, but even if there are, the inability to step away from the enter-combination-now portion to look for said clues is still a daunting problem.
This was also reviewed at The Rest of Your Mice.
Emily Short's Interactive Storytelling 
Yeah, I had a lot of trouble with this one too. I think we may have died in the same place, even? I was willing to cut it some more slack than you did (I may have eye-rolled at the quicksilver pickaxe, but I’ve seen stupider plot contrivances with less justification), but the typos and general lack of polish eventually turned me away, too. It’s a shame, because I actually liked the old-school gamebook* feel it had. That set itself apart from many other Twine games I’ve played.
* (i.e., Fighting Fantasy, Lone Wolf, et al.)
Yeah, for me one of these things by itself would not have been fatal — I might have powered through the puzzles if I’d been really into the story and setting, or I might have gone on with a dwarf-mines-setting game if the puzzles and typos etc. hadn’t made it a high-friction experience.
Dieties. I could not figure out how to pose the statues in relation to their ‘dieties’.
Definitely needed to be able ‘back away’ from puzzles that you couldn’t solve. Even with the walkthru, I couldn’t get the statues right :(
Am I the only one who found the poem on the desk? It gave the combination for the dowels pretty explicitly, and I think hinted about the inkwell too, although I found that without any trouble. The third line seemed to describe what the statues should be doing, but I couldn’t get a sensible combination to do anything, and the killing combination didn’t match the poem. I didn’t get any farther than that either.
It looks like the author made significant changes mid-comp based on this review. I think the theme it uses now is very sharp.