When I scaled this down, it was readable. While I haven't done a lot of the 8 pages at A6, that's a first. So many times people treat the pages as if they were still at A5 scale. Kudos there. It immediately also negates the normal complaint I would have about spending pages dedicated to legal elements in a tiny page count (I'd typically have those on the landing page or in fine print for a jam given the space limitations). Here though, it's basically in the same space as a single A5. So honestly it's more fit into a space most people are doing less with.
I'm not expert on the layout of A6, so just did a test check of 4 pages per sheet. I'll base my thoughts around that on how it is laid out. If I was just reading it in page setup strait down, I was thinking the map needed to come a little later, but then looking at it in the print layout, the next page pairs against it so I suppose it's a non-issue. It was just slightly jarring to have a map before I even had information about the setting.
Annex dungeon as a concept is solid. My own entry makes use of annex dungeons in one of it's elements. Making your entry into an annex dungeon means you avoid needing a huge amount of space spent on backstory and complex mapping. That said, it also means you have to make things a little more generic to allow it to annex in. As I read, I guess I will find out how that played out for you.
I like the teleportation coffin effect. That's interesting. Not sure yet how clear it will be to players but certainly fun. Having it alter time is interesting too, though presumably only one person fits at a time, maybe two? A bigger question I end up having is why the coffin slowly emerges from anywhere. If the players are retreading a dungeon back and forth I suppose that is an interesting little tidbit, but most of the time once they visit a spot, I find they rarely go back without good reason. I'd probably just end up having the coffin exist in a room somewhere and figure out a way to properly hint at how to go inside beyond an urge.
As I get to 'Exit' and 'Garden', I scroll back up to the map and sit looking at it for a minute. Okay. Things not indicated on the map itself. That's fine, setting elements. Did make me notice the map has a weird cut-off at the bottom that had me thinking something was there for just a second. There's probably a way to avoid that 'wait what' moment. Doubt there was time or testing to work it out though and it doesn't really hurt anything per se. Garden fruits are pretty clear about the risks there, so anyone eating that gets zero sympathy in my game.
At this point, I am realizing the bold is calling out elements well enough, but I think I would like some call-out boxes for things like the stat blocks on NPCs/Monsters. Separate them a bit from the other text of the document.
Funerary Panoply is not exactly a common word many people will likely know. At least based on my experience. When I have something where I think a word may confuse people, I tend to describe it along with the name/word. This is especially true since you bolded it, meaning it must be important as part of the description. Skipping ahead to the end, yes it is important. I would go ahead and put the information about the sentinel and panoply after the first entry. If I am referencing it, this is the most likely 'initial conflict' area. Since it is important to understand what a player is dealing with, having it here in entry 1 means the GM sees it immediately and can easily reference back to it just as easily as referencing to the end.
Area 2 is interesting in so much as the panoply associated with the sentinel here is in an entirely different area. Changes up the fight dynamics a little if they entered this direction. Area 3 offers a lot in a little space. A potential ally, a potential conflict with future encounters, a risk of misinformation, and a hope of useful information. This is good. It offers options for a lot of different player actions. Good choice.
The pit is where I have an exact opposite reaction to 3. It has all the same elements, but that's the problem. Aside from a slightly different personality, he is a carbon copy of Davke. Right down to no memory aside from his name, a single hint, and hatred of the other person. I think I would have varied this some.
Maybe he remembers the secret door along with a few tidbits of something, but is entirely unhostile towards Davke. Instead he acts defensively and confused, no longer even remembering why they are supposed to hate one another. Davke probably doesn't remember either, but is just feeling unbridled rage without remembering why. Something to consider. Right now they both remember their name, their hint, and how to coax the PCs into killing themselves. Carbon copy characters.
Disappearing gold is cruel. Not saying I wouldn't do it... but yeah. Depending on the player, some would be pretty upset over that. Not a trick I would pull too often lest they stop bothering.
I won't break down the rest since there's nothing else to really say. Lots of little places to move, but honestly I think the biggest problem is same-ness. I already mentioned the two NPCs being very similar, but almost everything is a mirroring. The enemies are all a single type with the only variation being if their connected panoply being in a different area or near them. There's a history between the NPCs, but even they apparently don't have memories of what it is outside of the lingering hostility.
It's not bad. I can't even pretend it's bad. It's a solid little dungeon. It's basic enough to fit in almost any game... but it's so basic in nature it lacks personality. NPCs aside, the only thing offering a history is the term 'Coffin of Enmezer', but I don't even see anything giving the players a way to know it has a name. If you plan to expand on this, that's where I would focus. Adding some variety to encounters and offering means of learning more about this interesting little sub-dungeon. Also including a clear reason to enter the coffin would help too. I don't think I have ever seen a PC try to get into a coffin. Loot one, yes. Climb in, no.
The 'urge to climb in' would scream trap to most sane players I think. I would probably choose to add a document the GM can drop elsewhere in their larger dungeon. Something that documents someone's attempt to enter, but not being able to get beyond the first sentinel. Perhaps he was with Davke and Gamul who climbed the tower and were supposed to open things up, but never came out! Bonus points if there's a fruit pit in his corpse's hand.
I like the idea of an annex dungeon and I don't dislike this usage. I think it needs some work to add a bit of history and personality more clearly. Not a bad little entry. Hopefully I've offered some useful things to consider and I look forward to seeing where this one places in the end.