Poland Company Registry Scraper — Official KRS/REGON APIs

Used by KYC analysts, M&A associates, and B2B sales teams who need live Polish company status, board members, and PKD activity codes without managing a KRS/REGON integration themselves.

Look up Polish companies via the official KRS (Krajowy Rejestr Sądowy — National Court Register) API — completely free, no key or signup required. Returns legal name, form, status, registered address, PKD activity codes, and board/representation. Optional REGON (GUS statistical register) enrichment resolves NIP/tax-ID numbers to full records, using your own free GUS key. Pay per result.

$4.50/1K records (Pay Per Event). First 10 results free. No proxy needed. No scraping ban risk. Government data source — authoritative, directly from the Ministry of Justice's KRS API.

Two data sources, two different access models

	KRS	REGON
Runs	Ministry of Justice	GUS (Central Statistical Office)
Auth	None — fully open API	Free key, but not instant self-serve
Lookup by	KRS number only (exact)	NIP, REGON, or KRS number
Search by company name	❌ Not available on this API	❌ Not available
Data richness	Rich: legal form, address, PKD codes, board/representation	Identity + address only

If you already have a KRS number (10 digits, e.g. 0000023302), you need nothing else — just krsNumbers and you're done, free.

If you only have a NIP (tax ID) or REGON number, KRS's own API has no way to resolve that — you need REGON, and REGON needs a key from GUS (see below).

Getting your REGON API key (only needed for nipNumbers / regonNumbers)

Unlike Companies House's instant web-form key, GUS does not offer self-serve production access:

1. Email regon_bir@stat.gov.pl with: your entity name, your own REGON, a contact person, and the IP address(es) your requests will come from (Apify runs use rotating cloud IPs — check with GUS how they want this handled, or run this actor from a fixed proxy/IP if they require it).
2. GUS says turnaround is "coordinated individually" — there's no published SLA. Budget a few business days.
3. Once issued, paste the key into regonApiKey.

To test the integration without waiting on GUS, tick useRegonTestEnv — this uses GUS's public sandbox key against a frozen, address-anonymized database snapshot from 2014. It proves the plumbing works; it will not return current real-world matches.

If you don't need NIP/REGON search, skip all of this — krsNumbers mode needs zero setup.

What you get

Field	Description
krs_number	KRS registration number (e.g. 0000023302)
nip	Tax ID (10 digits)
regon	Statistical register ID (9 or 14 digits)
registry_type	P (companies) or S (associations)
name	Full legal name
legal_form	e.g. SPÓŁKA AKCYJNA, SPÓŁKA Z OGRANICZONĄ ODPOWIEDZIALNOŚCIĄ
status	active, in_liquidation, bankruptcy, merged_or_transformed, terminated (see note below)
registration_date	Date first entered into KRS
last_entry_date	Date of the most recent register entry
address_line_1 / address_line_2	Street + house/flat number
locality	City/town
voivodeship	Region (województwo)
postal_code	Postal code
country	Country
website	Company website, when filed
pkd_main_code / pkd_main_description	Primary business activity (Polish Classification of Activities)
pkd_codes	List of secondary activity codes/descriptions
representation_method	How the entity can legally bind itself (e.g. "two board members jointly")
board_members / board_members_count	Current board/officers — [{name, function, suspended}]
registry_court	Court/authority behind the last register entry
source	krs or regon — which API produced this row
query	The input value (KRS/NIP/REGON) that produced this record
parse_confidence	Data quality score 0–1
warnings	Machine-readable quality flags

Privacy note: board members' PESEL (Polish national ID number) is never surfaced in the output, even masked — the government API itself already returns it pre-masked (e.g. 6**********), and we drop it entirely rather than pass through a partial identifier.

A note on status

Unlike UK Companies House, the KRS API has no simple active/dissolved field. We derive status from which register sections (dzial6 — mergers/liquidation/bankruptcy) are present on the entry:

• active — no liquidation/bankruptcy/merger section present.
• in_liquidation — entity is in formal liquidation.
• bankruptcy — bankruptcy proceedings are on file.
• merged_or_transformed — entity merged into / was absorbed by / transformed from another entity (common for banks/holding restructures — the entity may still be legally active).
• terminated — REGON-sourced records only, when DataZakonczeniaDzialalnosci (activity end date) is set.

This is an honest best-effort heuristic, documented here and in src/normalize.py. For hard compliance decisions, read registry_court / last_entry_date alongside status, or pull fullExtract: true for the complete historical record.

Modes

KRS direct lookup (free, default) — provide krsNumbers (e.g. ["0000023302", "0000025237"]). One record per number. Set registryType to S for associations/foundations, or auto to try both.

REGON enrichment (optional, needs your own key) — provide nipNumbers and/or regonNumbers plus regonApiKey. Each match returns identity + address (no board/PKD — those are KRS-only fields, left null on REGON-sourced rows, distinguishable via source: "regon").

Both together — combine any of the four list inputs in one run.

Use cases

• KYC / due diligence — verify Polish counterparty legal status, registration, and board
• B2B lead generation — enrich a list of Polish company names/NIPs with registry data
• Compliance screening — identify liquidated, bankrupt, or merged entities
• Beneficial ownership research — extract board/representation structures
• Market research — map PKD activity codes across a sector

Data sources

• KRS — Krajowy Rejestr Sądowy, the official Polish court register of companies/associations, maintained by the Ministry of Justice (Ministerstwo Sprawiedliwości). https://prs.ms.gov.pl/krs/openApi
• REGON — the statistical business register (Baza Internetowa REGON, BIR1.1), maintained by GUS (Główny Urząd Statystyczny). https://api.stat.gov.pl/Home/RegonApi

Pricing

Pay-per-result (PPE):

Event	Rate	Trigger
company-record	$4.50/1K	Every company record returned, from either source

No monthly subscription. No officer/board add-on fee — board data is included free whenever KRS returns it.

Worked examples:

Run	Records	Cost
100 KRS lookups	100	$0.45
1,000 KRS lookups	1,000	$4.50
500 NIP lookups via REGON	500	$2.25

FAQ

Do I need a proxy or API key? For krsNumbers lookups: no, nothing at all — the KRS API is fully open. For nipNumbers/regonNumbers: you need your own free REGON key from GUS (see "Getting your REGON API key" above — it's not instant).

Can I search by company name? Not on either official API — neither KRS nor REGON's DaneSzukajPodmioty supports free-text name search. You need a KRS number, NIP, or REGON number to start. If you only have a name, look it up once at wyszukiwarka-krs.ms.gov.pl (the Ministry's own portal, not part of this API) to get the KRS number, then feed that number to this actor for structured, repeatable, machine-readable data.

What if a KRS number doesn't exist? The API returns HTTP 404; that lookup is skipped and reported in the run's failedDetails, and it is not charged.

What output formats are available? JSON, CSV, and Excel — all via the Apify dataset.

Notes

• The KRS API is genuinely keyless and unauthenticated — verified live against https://api-krs.ms.gov.pl (2026-07-14).
• The REGON API is SOAP (BIR1.1), not REST — this actor handles the SOAP/WS-Addressing plumbing internally via zeep; you only ever provide a key.
• parse_confidence is 1.0 for records with a valid KRS/REGON number and name; it deducts for missing address, missing PKD, or fully empty responses.
• REGON's search-result field names in src/regon_client.py are documented from GUS's official technical instruction plus cross-referenced against independent open-source REGON clients; they could not be validated against a populated production response during development (only the empty 2014 test sandbox was reachable without a production key). Every field read is defensive — an unexpected field name degrades to null + a warning, never a crash. Run scripts/probe.py with a real production key to validate against live data before a large run.
• Live validation: run python3 scripts/probe.py (KRS, no key needed) or python3 scripts/probe.py --regon-test (REGON sandbox) to verify the integration end-to-end.

vs. Competitors

Feature	This Actor	HTML-scraping KRS actors (Store)	Paid KYC APIs
Data source	Official KRS REST API + optional REGON SOAP API	HTML scraping wyszukiwarka-krs.ms.gov.pl	Third-party aggregator
Proxy needed	No	Yes ($$$) — the portal is bot-protected (verified: returns HTTP 403 to direct requests)	No, but paid per call
Board/representation	Yes (from KRS dzial2)	Partial/unreliable	Varies
PKD activity codes	Yes (main + secondary)	Partial	Varies
parse_confidence	Yes	No	No
Cost	$4.50/1K	Similar + proxy COGS	Usually $10-30+/1K

Use with AI agents (MCP)

An agent calls this tool to look up Polish company registration status, address, PKD codes, and board members mid-conversation — e.g. "is KGHM Polska Miedź still active?", "who's on the board of KRS 0000023302?", or "resolve this NIP to a company name."

Point your MCP client at this tool:

{
  "mcpServers": {
    "apify": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.apify.com/?tools=bovi/poland-company-registry",
        "--header",
        "Authorization: Bearer <YOUR_APIFY_TOKEN>"
      ]
    }
  }
}

Minimal agent input (KRS lookup, no key needed):

{
  "krsNumbers": ["0000023302"]
}

With REGON enrichment (requires your own GUS key):

{
  "nipNumbers": ["7342867148"],
  "regonApiKey": "<YOUR_REGON_API_KEY>"
}

Example output row an agent can reason over directly:

{
  "krs_number": "0000023302",
  "nip": "6920000013",
  "regon": "39002176400000",
  "name": "KGHM POLSKA MIEDŹ SPÓŁKA AKCYJNA",
  "legal_form": "SPÓŁKA AKCYJNA",
  "status": "active",
  "registration_date": "29.06.2001",
  "locality": "LUBIN",
  "postal_code": "59-301",
  "voivodeship": "DOLNOŚLĄSKIE",
  "pkd_main_code": "07.29.Z",
  "pkd_main_description": "GÓRNICTWO POZOSTAŁYCH RUD METALI NIEŻELAZNYCH",
  "board_members_count": 5,
  "source": "krs",
  "parse_confidence": 1.0,
  "warnings": []
}

Reliability for agents: KRS data comes from the official Polish Ministry of Justice API (not HTML scraping), so results don't break on website redesigns. Every record includes a stable krs_number and/or regon/nip for de-duplication and joining across runs. Data quality is tracked via parse_confidence and machine-readable warnings.

Integrations

Built for KYC analysts, M&A associates, and B2B sales teams enriching Polish company status, board members, and PKD codes — the JSON/dataset output drops into the tools you already run, no glue code:

• n8n / Make / Zapier — trigger a run or pipe every new dataset item into 500+ apps (Google Sheets, Airtable, Slack, HubSpot, your database) with no code: n8n, Make, Zapier.
• Webhooks — fire your own endpoint the moment a run finishes, to push results straight into your pipeline (docs).
• MCP server — expose this actor as a tool to Claude, Cursor, or any MCP client so an AI agent can pull this data mid-conversation (guide).
• API & SDKs — fetch the dataset as JSON, CSV, or Excel through the Apify REST API or the Python / JS SDKs.

See all Apify integrations.

More scrapers from our toolkit

Building a data pipeline? These pair well with this one — each runs on your own Apify account with the same pay-per-result pricing, no subscription:

• UK Companies House Scraper
• Companies France