Home / Platform / Datahub

Outside data, on tap.

Datahub gives your agents curated external datasets, such as company filings, sanctions lists and commodity prices. We choose them, clean them and keep them up to date. Your agents search them in milliseconds, and every answer cites the dataset and the date it was current.

Datahub catalogall fresh
Companies and procurement48,212
Company filings21,480onupdated 2 h ago
Legal entity identities18,906onupdated 6 h ago
Government procurement7,826onupdated 2 h ago
Risk and regulation12,804
Sanctions & watchlists6,214onupdated 12 min ago
Regulatory change4,102onupdated 6 h ago
Vulnerabilities2,488onturned on just now
Markets, trade and climate3,190
Macro & economic indicators912onupdated yesterday
Commodity, metal & index prices604onupdated 12 min ago
Energy & agriculture388onupdated 6 h ago
Trade flows702onupdated 3 days ago
Weather & climate296offupdated 6 h ago
Research & patents288offupdated yesterday
Personal data minimized at ingestion12 categories · 64,206 records+4 more categories

A sample catalog. Each dataset is off until an admin turns it on.

How it works

Data your agents can trust, ready when they ask.

We pick each dataset and collect it from official APIs and bulk downloads, never scraping. Then we clean it, type it and update it on a schedule. When an agent asks a question, it gets an answer in milliseconds that names the dataset and the date behind it.

Cited answeranswered in 38 ms
Curated datasets
Sanctions & watchlistsupdated 12 min ago
Legal entity identitiesupdated 6 h ago
Company filingsupdated 2 h ago
Your agentvendor-screener

Can we onboard this new supplier?

The answer

No sanctions or watchlist match. The legal entity is active.

citedSanctions & watchlistsas of 2026-10-09 09:29
citedLegal entity identitiesas of 2026-10-09 03:41
Every answer names its dataset and dateExample figures
For your agents

What your agents get. From the first query.

Outside data your agents can use and your reviewers can check, without building an integration for each dataset.

Answers that cite their data

Every answer names the dataset it used and the date the data was current, so a reviewer can check it in seconds.

Kept fresh on a schedule

Each dataset updates on its own schedule.

Your rules, every read logged

A dataset stays off until an admin turns it on, and only the agents your rules allow can read it. Every read is logged.

One catalog for every solution

Turn a dataset on once and every solution you install can use it, from prospecting to KYC review.

Every dataset has a console

Try a query, read the schema, see which agents used it.

Each dataset shows its typed filters, its freshness, and the agents and runs that used it. You can query it yourself before an agent does.

Dataset console214 records in 6 ms
Government procurement awards7,826 records · updated 2 h ago
on
Schema
  • award_idkey
  • awarded_ondate
  • buyerentity
  • supplierentity
  • sectorenum
  • valuemoney
  • summarytext
  • updated_attime
deal-sourcersector: logistics · value ≥ 1M
  • pa-2026-04187logistics4.2M2026-03-12
  • pa-2026-03960logistics2.8M2026-02-27
  • pa-2026-03514logistics1.6M2026-02-09
  • +211 more records
Query it yourself before an agent doestyped · fresh · snapshotted
In the product

See what your agents read, and how fresh it was.

Freshness shows when each dataset last updated and when it updates next. The audit trail shows which agent read which dataset, when, and under which rule.

Freshnessall on schedule
  • Sanctions & watchlists
    updated 12 min agonext in 48 min
  • Company filings
    updated 2 h agonext in 22 h
  • Legal entity identities
    updated 6 h agonext in 18 h
  • Commodity, metal & index prices
    updated just nownext in 1 h
  • Trade flows
    updated 3 days agonext in 4 days
Each dataset shows its last and next updateolder versions kept by date

Freshness you can see

Each dataset shows when it was last updated.

In the marketplace

Solutions that come ready to use it.

These solutions use Datahub out of the box. Each read follows your constitution, and what they find is filed in your memory with the dataset and date it came from.

Sales & outreach

Prospecting

Finds target companies by sector, size and recent filings, then checks each against your do-not-contact rules before a draft is written.

Company filingsGovernment procurement
Sales & outreach

Account briefs

Pulls a company’s filings and legal entity record, with recent changes such as a new parent company, into a one-page brief before a call.

Company filingsLegal entity identities
Marketing

Market briefs

Sizes a market from company filings, procurement awards and trade flows, and cites the records it counted.

Government procurementTrade flows
Legal

Research memos

Gathers background on a counterparty from its filings and legal entity record before a contract review, and cites each one.

Company filingsLegal entity identities
The catalog

Chosen and cleaned. Kept fresh.

Your own systems of record, such as your CRM, ERP and document stores, connect to Celara directly. Datahub is for data from outside your walls. Personal data is minimized at ingestion.

48,212records in the sample catalog

Companies and procurement

Company filings, legal entity identities and government procurement, typed for search and enrichment, with full-text matching where the dataset supports it.

12,804records in the sample catalog

Risk and regulation

Sanctions and watchlists, regulatory change and vulnerabilities, kept fresh on a schedule.

3,190records in the sample catalog

Markets, trade and climate

Macro and economic indicators, commodity, metal and index prices, energy and agriculture, trade flows, weather and climate, and research and patents.

Example figures

Get started

Need a dataset we don’t have?

Tell us what you need.