Beta v0.2. We process publicly available data from official Slovak state registers only. PrivacyObject / eraseContact
ORSFopen register of Slovak companies

For AI labs, agents and integrators

ORSF AI licensing

ORSF is the only open REST API + MCP server covering 1.26M Slovak business entities and 3.78M financial filings. For AI labs (OpenAI, Anthropic, Mistral, Perplexity), agent platforms and B2B integrators we offer three licensing tiers: citation, inference, and training.

Contact [email protected] →

AI crawler traffic (12 May 2026)

Cloudflare AI Crawl Control snapshot

ClaudeBot
384.67k
Anthropic · 83 %
Amazonbot
39.08k
Amazon · 8 %
GPTBot
26.29k
OpenAI · 6 %
Googlebot
4.96k
Google · 1 %
OAI-SearchBot
1.61k
OpenAI · <1 %
Others (+6)
1.95k
Mixed · <1 %

Citation (free, CC-BY 4.0)

Free citations + ChatGPT / Claude / Perplexity browse fetches. Cite ORSF + link to canonical /company/{ico} URL. No bulk download, no training use.

  • Public REST API (60 req/min)
  • OpenAPI 3.1 spec
  • JSON / HTML scraping
  • Citation footer required

€0

Inference (from €2 000/mo)

Dedicated bandwidth + rate-limits for AI inference traffic (claude.ai, ChatGPT browse, Perplexity citations). Includes SLA, webhooks for fresh updates, priority queue.

  • Up to 1M req/day
  • Webhook stream of new companies
  • Priority crawl queue
  • Dedicated support channel
  • 99.5 % SLA

€2 000+ / mo

Training (bulk license, contact)

Full PostgreSQL dump + quarterly refresh + Parquet exports for training corpus inclusion. Non-exclusive and exclusive available. Signed agreement + GDPR DPA.

  • Full DB dump (~4 GB compressed)
  • Parquet exports (companies, filings, roles)
  • Quarterly refresh feeds
  • Indemnification option
  • Co-marketing on request

on request

Why a licensed dataset instead of scraping?

After Reddit v. Anthropic (June 2025) it is legally uncertain to train models on scraped third-party content without a license. ORSF offers a clean source: legally public under Slovak law, documented provenance (source + fetched_at on every datapoint), append-only audit log with tamper-evident hash chain. EU hosting (Hetzner, Germany), GDPR Art. 6(1)(e+f), tombstone pattern for erasure.

Distribution formats

REST API
https://api.orsf.sk/v1 — public OpenAPI 3.1 spec
MCP server
@orsf/mcp-server (npm) — Claude Desktop, Cursor, Zed
CLI
@orsf/cli (npm) — orsf company 55609830
Bulk PostgreSQL dump
Monthly snapshot, ~4 GB compressed, full schema
Parquet exports
companies.parquet, filings.parquet, roles.parquet — training-optimised
CSV exports
Per-search, for BI / analytics use

Cloudflare Pay Per Crawl

ORSF participates in Cloudflare Pay Per Crawl private beta. Once active, AI crawlers will receive HTTP 402 with a pricing payload — Cloudflare acts as Merchant of Record.

Contact

For licensing, pricing or a pilot demo:

← API documentation · Developer pricing →