The data nobody else archives, served over one API

Live coverage, as of the last stats refresh: 1381 scraper modules running against 1574 tables in the production database — 728 of those scrapers run daily-or-finer(hourly to every-few-hours; the rest weekly/monthly where the underlying source itself is that slow) — out of 1,543+ catalogued across 24 domains. Coverage grows every night, and this line only ever reflects what is actually running, not the full catalog.

A sample of the live catalog below — weighted toward daily-or-finer feeds, the ephemeral data that makes this a live sensor network rather than a report archive:

Representative of the 1,543+ data feeds catalogued across 24domains — browse every domain below, or the full live catalog further down the page.

The value of the library is that most of these feeds are ephemeral and archived nowhere else. COT reports are available to every desk; a page that changes daily and is silently overwritten, with no history retained at the source, is not. These are data feeds, not trading signals — only research that clears the five-stage validation process on /aboutwould ever be marked “validated,” and to date none has.