Data Reference
Every dataset, documented.
We do not ask you to take coverage on faith. Every dataset below lists its exact row count, coverage window, update cadence, the source filing it is built from, its key fields and an example response -- measured against the live production cluster. The numbers you query are the numbers we show.
From filing to data
The messy filing, made machine-ready.
SEC documents are dense, nested, and inconsistent. Your agents never touch them — we hand back the clean, typed result: one field per value, the entity resolved, and a link to the exact filing it came from.
13F Holdings
Every institutional position from Form 13F — shares, market value, portfolio weight, and the quarter-over-quarter action, consolidated per manager.
Every dataset, one row each. Click any to open its key fields and an example response.
Ownership & Insider
Who owns what, and who is buying or selling their own stock -- straight from SEC ownership filings.
Corporate & Governance
Material events, fundamentals, activist stakes and pay -- the corporate disclosure record.
Market Structure & Government
Settlement failures, short-squeeze watch lists, and the trades of the people who write the rules.
Macro & Economic
Rates, the curve, the Fed balance sheet, inflation, positioning, banks and energy -- the federal data layer.
Row counts and coverage windows measured against the production cluster, 2026-06-03. Example payloads show the response shape (field names and value types); they are representative schemas, not live quotes.
Methodology
Why you can trust every number.
Source-traced
Every SEC row carries the accession it came from. Nothing is editorial; you can open the original filing for any number on the site.
Normalized & deduped
Share classes (GOOGL+GOOG) are consolidated, and filing entities that restructure (e.g. a manager that splits into sub-entities) are merged back to one canonical manager so portfolios stay continuous.
Test-validated
49 automated data-quality tests (uniqueness, not-null, range, reconciliation) gate every publish. A filing that fails validation never reaches the API.
Replicated & fresh
Stored across 3 independent regions with daily freshness checks. The same data the API and MCP serve is the data you browse here -- no stale mirror.
Query it yourself.
Every dataset is one REST call or one MCP tool away. Start free, no card required.