Exercise 2: What Each @jsonrender/* Package Actually Does — Possible Solution
====================================================================

@jsonrender/core is the engine itself - the part of the pipeline
responsible for actually taking a spec and driving the render process,
independent of any specific UI library. @jsonrender/react-schema is
the schema layer, defining what shape a valid spec is allowed to take
in a React context - the rules a spec has to follow before it can be
trusted. @jsonrender/shad-cn is the binding layer specifically for
Shadcn - it connects the abstract component types described in the
schema to real, concrete Shadcn component implementations.

The project needs all three separately, rather than one combined
package, because each one handles a genuinely different, swappable
concern. The core engine doesn't need to know anything about React or
Shadcn specifically; the schema layer defines valid structure without
caring which UI library eventually renders it; and the Shadcn bindings
are what could, in principle, be swapped out for a different component
library's own bindings without having to rewrite the core engine or
the schema rules at all. Splitting them keeps each piece focused and
independently replaceable.

ANSWER: @jsonrender/core drives the render process itself,
@jsonrender/react-schema defines what a valid spec looks like, and
@jsonrender/shad-cn connects the schema's component types to real
Shadcn implementations. They're kept separate because each handles a
distinct, independently swappable concern - most notably, the Shadcn
bindings could be replaced with a different component library's own
bindings without touching the core engine or schema layer.

WHY THIS WORKS AS AN ANSWER
------------------------------
This correctly assigns a distinct real role to each of the three
packages and explains the architectural reason for keeping them
separate (independent swappability) rather than just restating their
names.