alibaba/GraphScope — reverse-engineered prompt

Build me a one stop graph computing platform like GraphScope that lets me work with very large property graphs from Python without a lot of setup. I want to be able to load graph data with labels and properties, run interactive graph queries, do graph analytics, and try graph neural network style workflows in the same place. Please make the first experience simple in standalone mode on one machine, but keep it ready to run on a Kubernetes cluster for bigger jobs later.

I’d also like a clear getting started experience with an easy install path, plus a notebook or example script that shows a full workflow on a citation style network for node classification. It should walk through loading the graph, exploring it interactively, running a few analytics steps, and then using both graph structure and node features for the prediction task. Good docs and a smooth demo matter more to me than polishing every edge case, so look up the current docs online if you need to.