I think I finally have a concept I like enough not to throw it out again! First, a short overview of the three ideas I have thrown out because they are too complicated for one person in 48 hours or because I realized I didn't actually like them:

Space station loses signal

The player was to be on a rotating mining station where the whole mining and refining process only works correctly by following the rhythm from the central processing stations signal beam. But the dish gets knocked askew and instead catches some other signal and all the machinery now runs wildly forward and back based on the new (wrong) signal. You'd have to navigate to the other side of the station to realign the radio dish.

Too many interlocking systems for one person

Traffic control laser canon

You shoot everyone in town who turns or changes lanes without using their turn signal with a giant laser canon to blow them up.

Simple, but not worth more than a short chuckle. Not enough substance

Traffic conductor by way of rhythm game

What it says on the tin.

Doable, but too much work in generating the actual rhythm parts for a genre I don't like all that much.

So, what now?

For my fourth and final idea, the idea is to put the player into a labyrinth that constantly shifts, and they can only find out where the way out is by using a signal strength reader that hones in on the shortest path to the signal source back on the surface. The complications would then come from having to find ways to get around closed iron grates and collapsed bridges that are the shortest path for the signal, but not the player.

Wait, what did you actually end up using to make this?

I've settled on zig and raylib. It's everything I loved in pygame when I first started, except it also makes getting started on 3D extremely easy if you want/need it.

I just hope all the basic scaffolding I've already written saves me at least some of my lost time. If this doesn't work out, I can always pick up #2 again later, that one should be quick.