Public Company Registries — 11 Countries

Look up official company data from European public registries via Python scripts (direct REST API) or Apify actor fallback (scraping when no API exists).

Prerequisites

• Python 3 (scripts use stdlib — no pip dependencies for most countries)
• Apify access (only for DE, UK, PL financials, CZ justice.cz, RO fallback) — preferred: apify CLI (npm install -g apify-cli && apify login); fallback: Apify MCP connector (call-actor tool). CLI is faster and preferred when both are available.
• NL: KVK_API_KEY env var (register at developers.kvk.nl — see REGISTRATION/NL.md)
• PL GUS: GUS_API_KEY env var (email regon_bir@stat.gov.pl)
• SE: Bolagsverket requires registration (see REGISTRATION/SE.md)
• HR: Sudski registar requires OAuth registration (see REGISTRATION/HR.md)

${CLAUDE_PLUGIN_ROOT} is the plugin's root directory (where .claude-plugin/ lives). It is resolved automatically by Claude Code when the plugin is installed, or set to the --plugin-dir path during development.

Workflow checklist

Copy this and tick boxes as you progress:

Task Progress:
- [ ] Step 0: Verify prerequisites — run `python3 --version` (required); for Apify-dependent registries: try `apify --version && apify info`, or check for `call-actor` MCP tool; if neither, tell user to install apify CLI or Apify MCP connector
- [ ] Step 1: Identify country + identifier type (IČO/KRS/CUI/LEI/company name)
- [ ] Step 2: Run the lookup command per country table
- [ ] Step 3: Interpret results (key fields vary by source)
- [ ] Optional: Apify fallback for scraping-based registries
- [ ] Optional: Cross-reference with EU-level sources (GLEIF LEI, ESMA bonds, EBA)

Quick Reference — Lookup by Country

All paths relative to ${CLAUDE_PLUGIN_ROOT}/skills/apify-public-registries/.

By Data Type

Step 1: Identify country + identifier

• Check ${CLAUDE_PLUGIN_ROOT}/data/companies.json -> identifiers.registry_ids for existing IDs.
• If identifier unknown, start with EU GLEIF by name — returns LEI + registered-as numbers usable in country-specific lookups.

Identifier types by country:

• CZ/SK: IČO (8-digit)
• PL: KRS (10-digit zero-padded) or NIP (10-digit)
• DE: Company name keyword
• UK: Company name
• NL: KVK number (8-digit)
• RO: CUI (numeric)
• HR: OIB (11-digit) or MBS
• SE: Organisationsnummer (10-digit, e.g. 559124-6847)
• EU: LEI (20-char alphanumeric) or company name

Step 2: Run lookup

Execute the command from the country table. Scripts are at: ${CLAUDE_PLUGIN_ROOT}/skills/apify-public-registries/reference/scripts/<CC>/fetch_all.py

Example — look up a company in Czech ARES:

python3 ${CLAUDE_PLUGIN_ROOT}/skills/apify-public-registries/reference/scripts/CZ/fetch_all.py lookup 25099345

Example — find ownership chain via GLEIF:

python3 ${CLAUDE_PLUGIN_ROOT}/skills/apify-public-registries/reference/scripts/EU/fetch_all.py lookup_lei 31570048XH84U51GGT05

Example — look up InPost in KRS:

python3 ${CLAUDE_PLUGIN_ROOT}/skills/apify-public-registries/reference/scripts/PL/fetch_all.py lookup 0000536554

Batch fetch (all companies in a country):

python3 reference/scripts/CZ/fetch_all.py          # all CZ sources (ares, dph, cnb_banks, cnb_oam, justice)
python3 reference/scripts/PL/fetch_all.py          # all PL sources (krs, biala_lista, gus, financials)
python3 reference/scripts/EU/fetch_all.py          # all EU sources except EBA (gleif, ted, esma, eurostat)

Step 3: Interpret results

Key fields vary by source:

Apify fallback

When no direct API exists, use Apify CLI with a specific actor or apify/website-content-crawler:

apify call apify/website-content-crawler \
  --input '{"startUrls":[{"url":"https://or.justice.cz/ias/ui/rejstrik-$firma?ico=25099345"}],"maxCrawlPages":1}' \
  --user-agent apify-awesome-skills/apify-public-registries

Apify-dependent registries:

Note: DE, UK, and RO scripts use mcpc CLI internally (not apify CLI).

Dependencies

Python stdlib only (no pip)

CZ (ares, dph, cnb_banks, cnb_oam, justice), SK (orsr, finstat), PL (krs, biala_lista, gus), EU (gleif, esma, ted, eurostat), ESG (ets, climate_trace), RO (anaf)

Requires Apify + APIFY_TOKEN

DE (handelsregister), UK (companies_house), PL (financials via eKRS), CZ (justice.cz sbírka listin via scraping), RO (listafirme.ro)

Requires API key registration

• NL: KVK_API_KEY — register at developers.kvk.nl (REGISTRATION/NL.md)
• PL GUS: GUS_API_KEY — email regon_bir@stat.gov.pl (test key abcde12345abcde12345 exists but has no real data)
• SE: Bolagsverket — register at portal.api.bolagsverket.se (REGISTRATION/SE.md)
• HR: Sudski registar — register at sudreg-data.gov.hr, OAuth2 flow (REGISTRATION/HR.md)

Critical gotchas

• CZ ARES is the best single starting point for Czech companies — aggregates data from multiple source registers.
• EU GLEIF is the best cross-country starting point — maps LEI to registered-as numbers usable in country lookups.
• RO ANAF API is offline since March 2026 — lookup returns errors. Use listafirme.ro (Apify) as fallback.
• SK ORSR uses HTML scraping with windows-1250 encoding and regex parsing — fragile, may break if page layout changes.
• SK FinStat is also HTML scraping — may block automated access.
• PL Biała Lista has a hard rate limit: 100 queries/day (search method) or 5,000/day (check method).
• PL GUS silently returns fake data when GUS_API_KEY is unset — test server returns plausible-looking records (ul. Test-Krucza, Kraków-Podgórze) for any valid NIP. Production key required: email regon_bir@stat.gov.pl. Always inspect first GUS record to confirm prod vs test.
• DE + UK scripts use mcpc CLI (not apify CLI) — require authenticated @apify session.
• UK Companies House has a free API at api.company-information.service.gov.uk, but the script uses Apify scraping instead.
• CZ Justice.cz bulk datasets are 13GB+ XML/CSV. The CKAN API returns a dataset list, not per-company data.
• CZ ČNB OAM is only relevant for emitents (listed companies, banks) — uses Oracle BI XML export.
• ESG data is country-level, not company-level — useful only as contextual benchmark.
• ESAP (centralized EU ESG/financial database with API) launches CSRD data January 2028.
• EBA Transparency CSVs are 100MB+ each — run EU/fetch_all.py eba separately, not as part of normal batch.
• TED has low value — most portfolio companies are private. Winner-name only in eForms notices (2024+).
• Batch mode (fetch_all.py without arguments) reads lookup_targets.json from the script's directory — these are per-country ID→name mappings, NOT copies of data/companies.json. For single-company lookups, pass the identifier directly.

Reference

• data/registries.json — machine-readable: 11 countries, 27 registry entries
• REGISTRATION/HR.md, REGISTRATION/NL.md, REGISTRATION/SE.md — per-country API access guides
• reference/european-company-data-institutions.md — full narrative catalog (413 lines, CZ + PL + EU institutions)
• reference/scripts/{CZ,DE,ESG,EU,NL,PL,RO,SK,UK}/fetch_all.py — per-country lookup scripts