Thanks for your helpful replies!
I will definitely start getting involved in the Discord soon and will try contacting the musicians you have mentioned.
And yes, I have a decently functional production framework for 360-degree nodes already figured out in Unity, that I figured out while developing the first few areas of Miniature Multiverse (That project is still incomplete but steadily moving forward - I would now expect it to be done some time around February or March 2020 at the rate things have been going.)
I actually tried Dagon at one point when that was first emerging, and I once attempted a big game with Adventure Maker, a project I discontinued almost a decade ago because even then I saw 'the writing on the wall' and realized that the engine probably had no long term future. The developer stopped updating the engine in something like 2007 or 2008, and it is surprising actually that anyone still uses it given how many of the advanced or plugin features in its output become increasingly broken for each and every version of Windows released after XP. I even tried messing with a panoramic VR tour engine once, that was not even a game engine but I figured out sneaky workaround ways to script simple puzzles and interactions in it with varying game states.
Eventually though I realized Unity could do this and I started testing that with a little proof of concept... and I knew it could work.
Doubly cool is that I think it can even be done stereo VR, ie with convincing eye separation depth, under certain conditions. I have successfully tried a couple of methods that both work for that. One - which is a little nicer looking - requires the rendered scene source be a 3d scene rendered in stereo (two 360-degree rendered versions of every node and transition slightly offset) the other is a weird set of tricks using a custom designed equirectangular displacement map applied to reshape and distort the node sphere, so that from the center the illusion of varying and detailed depth is possible, even when the panorama is not 3d rendered and is primarily something else like miniature photography. (as is of course the case with about 75% of the imagery in Miniature Multiverse.)
I am toying with a range of possible rough ideas for this game jam but without the secondary theme revealed I am hesitant to fully design any storyline, or puzzles that fit it well, mostly I am waiting for the thing to start before identifying where exactly I want to go with this. Designing and coming up with neat ideas, IMO, is fun but not anywhere near as time consuming as a high quality execution, that makes those ideas into a good, polished experience.
I will say I have a couple of interesting loosely thought-out stories woven around notable myths / legends... that lead to some intriguingly weird reinterpretations of stories 'everybody already knows' and suggests an alternative explanation or context for the story that is arguably even stranger and maybe more interesting than the straightforward myth was originally in the first place.
I like the sense of having the player dumped into the storyline initially having no clue where it is headed. Keep them guessing, hit them with an interesting reveal or two - roughly at the midpoint and near the ending - to keep them off guard and intrigued as it all progresses, unsure of certain aspects of the story and trying to make sense of the information they have learned, trying to answer their questions through exploring, observing, and solving puzzles. Basically the prototypical feeling of confused fascination players felt when dumped onto the dock on Myst Island back in '93, before the thing became less mysterious and more... known and quantified.
Part of the trick is to do this in some ways I haven't already committed to doing elsewhere in any of my other stories (past or currently in development). I think I may have the seed of something sort of fresh though, a bit different than what I have already tried or planned elsewhere. A rough story structure which could work, could become something really cool if I pull it off competently and resolve the specific details and logic, delivering well on the inherent potential within its premise.
But who knows, I may botch it when time starts to run out. Can't allow feature creep here on a three week effort and need to keep things reasonably simple to meet the deadline. :/