See the linked comment thread for more context. The TL;DR is that the existing endpoint for listing timeseries schema is probably unhelpful and should be deleted. It was built when we had no consumers and thought we might export raw data by timeseries. At this point, we'll likely move towards exposing data about resources, which will be internally derived from one or more timeseries in the database. But the raw schema will probably not be visible in the public API.
Originally posted by @ahl in #1329 (comment)