KYB Company Verifier — Cross-Registry Business Verification API

Verify whether a company exists and is in good standing across multiple official business registries in one keyless call. Give it a company name and it fans out — in parallel — to France (INSEE/RNE), the GLEIF LEI database, and US state registries, returning one normalized record per registry with: does it exist, legal status, jurisdiction, identifiers (SIREN / LEI / state ID), officers or registered agent, address, source link, and a match-confidence score so you can auto-accept or route to manual review.

Built for KYB / AML onboarding, merchant/vendor due diligence, and AI agents that need a grounded "is this a real company?" check.

Who it's for

• KYB / AML & compliance teams grounding onboarding against primary sources.
• Merchant & vendor onboarding / RevOps auto-verifying business identity before approval.
• Fintech / payments risk screening applicants at signup.
• AI agents & builders that need a callable company-existence tool (see MCP note below).

What it does

• Cross-registry fan-out from a single company name (parallel, keyless).
• Normalizes each hit to a stable schema with a status enum (active / inactive / dissolved / unknown).
• Scores name match (token similarity, legal-form aware) so DANONE SA == Danone == confidence 1.
• Emits an explicit exists: false row when a registry returns no match — negative evidence is a first-class result.

Supported registries

Code	Source	Jurisdiction
FR-INSEE	recherche-entreprises.api.gouv.fr (INSEE + RNE)	France
GLEIF	api.gleif.org LEI records	Global
US-CO,US-CT,US-NY,US-OR,US-PA,US-DE	State open-data registries	United States

Example input

{
  "companyName": "Danone",
  "country": "FR",
  "maxPerRegistry": 3
}

Query everything (leave country/registries blank), or target specific registries:

{
  "companyName": "Acme Holdings LLC",
  "registries": ["GLEIF", "US-CO", "US-NY"],
  "maxPerRegistry": 5
}

Output fields

Each dataset item is one registry result:

Field	Description
input_name	The name you queried (echoed for joins).
registry	FR-INSEE, GLEIF, or US-<state>.
jurisdiction	FR, US-CO, GLOBAL, etc.
exists	true if a match was found, else false.
matched_name	Registered legal name of the match.
entity_id	SIREN (FR), LEI (GLEIF), or state entity/filing number.
status	Normalized: active / inactive / dissolved / unknown.
address	Registered / head-office address.
officers	Directors, officers, or registered agent (joined).
match_confidence	Name similarity in [0,1] (1 = exact; 0 for not-found).
source_url	Link to the official record for auditing.
retrieved_at	ISO-8601 UTC timestamp.

Use as an MCP tool

This actor is callable by AI agents (Claude, Cursor, etc.) via mcp.apify.com. The clean, field-level input/output descriptions make it a reliable company-verification tool an agent can chain into a KYB/AML workflow: pass a companyName, read back exists, status, entity_id, and match_confidence per registry.

FAQ

How do I verify a company exists by name? Pass companyName; each registry returns exists plus a match_confidence so you can threshold auto-accept.

Is this a KYB / AML company verification API? Yes — it grounds business identity against official French, global (GLEIF), and US state registries.

How do I get a company's SIREN or LEI from its name? The entity_id field returns the SIREN for French matches and the LEI for GLEIF matches.

Does it tell me if a company is dissolved or inactive? Yes — status is normalized to active / inactive / dissolved / unknown across registries.

Do I need an API key? No. All sources are keyless open data / public APIs.

Notes & limits

• A registry that errors is skipped (logged) rather than reported as exists: false — an outage is not proof of absence.
• Delaware exposes licensed businesses (not the corporate registry); treat US-DE as a licensing signal.
• Name matching is fuzzy; always gate automated decisions on match_confidence.