If it runs fast and low resource, it could be a nice testbed for agent architectures. I'm thinking old-school planners, genetic programming, ML techniques, etc.

Especially integrating planners or logic-based approaches with ML models. The old architectures for StarCraft were great at building and micro-managing units but terrible at assessing or adapting to enemy strategy. Hybrid architectures can address that.

Games like yours, esp different styles or requirements, can be used to prove how well the techniques generalize.

that's a good idea, I could definitely see this and games like this as agent benchmarks, like an ARC-AGI but it rotates amongst strategy / long-horizon sandbox games, so we don't have benchmaxxing..

It's a good direction though, I'll take this into consideration in future versions :D