For this PCG setup I built a simple global system that lets me plug in different PCG graphs and enable or disable them as needed. The core of it is a custom Data Asset with two fields: Graph and Use. Graph holds a PCG graph reference, and Use is just a boolean that decides whether that graph should run.

The idea is to make the system modular and easy to extend. You can add as many graph entries as you want, whether they are for cliffs, grass, paths, or any other world-building feature, and control them from a single place without having to rewire the whole setup.

Internally the logic is very straightforward. One graph reads the Data Asset and pushes its entries into a loop. The second graph is the loop itself: for each entry, if Use is enabled it runs the corresponding PCG subgraph, and if it is disabled it simply skips it. It is a small piece of logic, but it gives a lot of flexibility and keeps the system easy to manage.