terpstra-209 fixes a production build blind spot
The fork restores the type information the frontend needs from the backend, preventing a deployment-time failure.
This is a small but practical reliability fix for teams deploying the application through its container setup. The frontend can again see the shared descriptions of data coming from the backend, so production builds no longer fail when those references are checked.
The change also keeps the frontend and backend more tightly linked during the build process. That solves the immediate deployment issue, though a cleaner long-term approach would give both sides a dedicated shared source for those definitions.
Spotted something wrong? Or know the PR text has fresher detail than the writeup above?