Data ingestion

Fetches batches of names from Wikidata and persists them to the configured database (D1 in production, SQLite locally).

Use this tool before enabling the public API, so requests can serve from cached tables instead of live SPARQL queries.