Every source in the substrate, with its provenance.

The Vlada substrate is the agent-ready layer of US healthcare cost data. This catalog lists each source — what it is, its grain, where to query it, what it does not yet cover — and the tools registered against it. Generated from catalog.json by scripts/build_catalog.py.

Context docs and provenance refs are repository paths (e.g. docs/context_library/…), not links — read them in the repo for the full source contract.