Datasets

Every dataset family, its dataset ids and the route that serves it.

A dataset is one kind of row with one schema and one route. Datasets are grouped in families; each family has a guide that explains the data and its caveats, and a reference page with every route's parameters. GET /v1/datasets returns the same list from the live API, with the path of each.

How to read the state of a dataset (meta.status, 503 terms_unread) is in Freshness and availability. These pages do not say which datasets currently hold data, because that changes as sources are switched on.

Families#

Family	Dataset ids	Guide	Reference
Market data	`finance.market.candles_1m`, `candles_1h`, `candles_1d`, `instruments`, `tickers_latest`, `funding_rates`, `open_interest`	Market data	Market
Macro	`finance.macro.calendar_events`, `series`, `observations`; `finance.macro_derived.series`, `observations`	Macro	Macro
Filings and trades	`finance.filings.companies`, `filings`, `company_facts`, `fundamentals`, `insider_transactions`, `fails_to_deliver`, `holdings_13f`, `news_8k`, `congress_trades`	Filings	Filings
Taiwan equities	`finance.equities_tw.securities`, `daily_close`, `valuation_daily`, `institutional_daily`, `margin_daily`, `monthly_revenue`, `statements`, `indices`, `sources`	Taiwan equities	Taiwan equities
Statutes	`legal.statutes.tw_moj`, `jp_egov`, `uk_legislation`, `ca_laws`, `eu_eurlex`, `us_ecfr`, `de_gii`, `au_federal` (and `sg_sso`, `cn_flk`, see the guide)	Statutes	Legal
Name lists and legal documents	`legal.regulatory.ofac_sdn`, `un_sanctions`, `us_exclusions`, `fsc_news`; `legal.contracts.tw_standard_forms`	Regulatory and legal documents	Legal
Health	`health.drugs.fda_`, `health.devices.fda_`, `health.drugs.ema_medicines`, `health.providers.nppes`, `health.trials.clinicaltrials_gov`, `health.statistics.who_gho`, `health.statistics.eurostat_mortality`	Health	openFDA, registries
On-chain	`onchain.defi.protocol_tvl`	On-chain	On-chain

Choosing a route#

Time series (candles, funding, open interest, indices, prices): filter by id and a from/to window.
Catalogues (instruments, series, companies, securities): small lists you can cache; page with cursor.
Records (statutes, trials, medicines, filings): list with filters, then fetch one by its id.
Search: name lists and provider registries are searches, deliberately not bulk enumerations. Statute titles can be filtered by substring; there is no full-text or semantic search over statutes on the data routes (retrieval over judgments and statutes is a separate service, see Statutes).

What is not offered#

No streaming: no WebSocket or server-sent events on the data API. Everything is HTTP request/response.
No browser access to the data API (no CORS). Call it from a server or an agent.
No order books, individual trades or liquidations in the market data, only candles, tickers, funding and open interest.
Statutes are the current version only; old versions are not kept or served.
Sources whose publishers offer no official API or licence for redistribution are not collected (for example Singapore Statutes Online and the Chinese national law database).

Source: content/datasets/index.md