Call this before writing any aggregate query over a DuckLake table. A declared measure is the canonical definition of that number, and reproducing it yourself will silently disagree with it (a revenue measure typically excludes refunds or test rows). Filter by table for one table’s declarations, or by path_prefix (e.g. f/analytics) for everything declared under a folder; omit both to browse the whole catalog. Results are keyset-paged: a full page may mean more remain, so continue with the cursor_* params rather than assuming a measure does not exist. Use each returned expr verbatim, and when a measure has a filter write it as expr FILTER (WHERE filter) so measures with different predicates can share one GROUP BY. If a number you need has no declared measure, write your own aggregate as usual. Results are limited to declarations whose producing script the caller can read.