Openregistry

Created By
sophymarine2 months ago
Overview

openregistry

Real-time queries. Raw data. Raw documents. Synced directly from official registries — no intermediaries.

openregistry is a free remote MCP server — a platform by sophymarine — that connects AI agents directly to official national company registries as typed, structured tools. Search companies, fetch profiles, list filings, retrieve officers and beneficial owners, trace directors across companies, and download raw filing documents — all without leaving your AI client.

Every call proxies directly to the official registry in real time. There is no third-party data warehouse, no nightly scrape, no AI interpretation between you and the official record.

Hosted endpoint: https://openregistry.sophymarine.com/mcp


Why openregistry

  • Real-time. Every tool call is a live query to the upstream registry API — not a cached snapshot, not a third-party aggregator, not a nightly scrape. Pass fresh=true to bypass even the short-lived performance cache.
  • Raw data. Company profiles, officer lists, shareholder registers, beneficial-ownership entries, charges, and filing metadata come back with upstream field names preserved. No opinions. No AI interpretation between you and the record.
  • Official source. Data comes from the statutory registry of record for each jurisdiction — not a scraper third-hand.
  • No intermediaries. No third-party data warehouse, no nightly scrape, no translation layer between your AI and the registry.
  • Free for anonymous use. No API key, no account, no installation — add the server URL and it works. Sign in (passwordless, email magic link) for higher rate limits and multi-country search fan-out. See Tiers.

Jurisdictions covered

RegionRegistries
UK & Crown DependenciesGB Companies House, IM Isle of Man, KY Cayman CIMA
EUFR RNE, DE Handelsregister, IT InfoCamere (via BRIS), ES BORME, NL KVK¹, BE KBO, IE CRO, PL KRS, CZ ARES, FI PRH, CY DRCOR, LI Handelsregister, MC RCI
NordicsNO Brreg, FI PRH, IS Skatturinn
SwitzerlandCH Zefix + SHAB/SOGC delta stream
North AmericaCA (federal CBCA, CA-BC OrgBook, CA-NT), US-NY, US-CA, US-FL, US-CT, US-PA, US-CO, US-OR, US-IA, US-OH
Latin AmericaMX PSM, BR CNPJ²
Asia-PacificAU ABR, NZ Companies Office¹, HK CR, TW GCIS, KR OpenDART, MY SSM, ID AHU, IN MCA
CISRU FNS + GIR BO + Interfax

¹ listed in registry, coverage incomplete · ² identifier-only lookup

list_jurisdictions returns the authoritative, up-to-date capability matrix — which tools are live per country, which are paid-tier-only, which return 501 alternative_url pointing at a paid upstream.

Tools

ToolWhat it does
search_companiesFind companies by name. Supports single-country direct search and user-confirmed multi-country fan-out.
get_company_profileFull company profile with every upstream field preserved in jurisdiction_data.
get_officersCurrent and historical directors / secretaries / officers.
get_shareholdersShareholders / quota-holders where the registry publishes them.
get_persons_with_significant_controlBeneficial-ownership register entries (UBO).
get_chargesRegistered mortgages and security interests.
list_filingsFiling history with document IDs and category filter.
fetch_documentRaw filing bytes (XHTML / PDF / XML) or source URL.
get_document_metadataDocument formats and sizes before downloading.
get_financialsAnnual financial-statement filings with normalised fiscal-period shape.
search_officersFind people by name across all company appointments.
get_officer_appointmentsEvery company a person has been appointed to.
list_jurisdictionsDiscover the live capability matrix per registry.
aboutCompact capability + pricing summary.

The hosted API reference documents per-jurisdiction coverage, identifier formats, and 501 routing for paid-upstream fallbacks.

Connect

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "openregistry": {
      "url": "https://openregistry.sophymarine.com/mcp",
      "transport": "http"
    }
  }
}

Cursor

Add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "openregistry": {
      "url": "https://openregistry.sophymarine.com/mcp"
    }
  }
}

Claude Code

claude mcp add --transport http openregistry https://openregistry.sophymarine.com/mcp

Cline (VS Code)

Settings → Cline → MCP Servers → Add:

{
  "openregistry": {
    "url": "https://openregistry.sophymarine.com/mcp",
    "transport": "streamable-http"
  }
}

See llms-install.md for automated installs by LLM-driven clients.

Anything else speaking MCP

Streamable HTTP transport per MCP spec 2025-06-18. OAuth 2.1 authorization flow for the authenticated tiers (Dynamic Client Registration per RFC 7591 — no API key to paste).

Tiers

TierPriceRate limitMulti-country search fan-out
Anonymousfree20/min per IP3 countries / 60s
Free (signed in)free30/min per user3 countries / 60s
Pro$9/mo180/min per user10 countries / 60s
Max$29/mo900/min per user30 countries / 60s
Enterprisecontact sales3000/min per userunlimited + source URLs included

All tiers receive the same unified schema and full raw upstream data. Enterprise adds synthesised source-URL / registry-name / data-license fields.

Full pricing and tier details: openregistry.sophymarine.com/tiers.

Examples

Look up a company and fetch its latest accounts:

"Search Companies House for Monzo Bank, get their full company profile, and show me their most recent accounts filing."

The assistant will chain search_companiesget_company_profilelist_filings(category='accounts')fetch_document and analyse the filing inline.

Trace a director across every company they've run:

"Find every UK company where Elon Musk has been a director, and show me which ones are still active."

search_officers(GB)get_officer_appointments(GB).

Beneficial ownership analysis:

"For company number 00000006 (Pilkington), who are the persons with significant control, and what mortgages are registered against them?"

get_persons_with_significant_control(GB) + get_charges(GB).

Figure out the country first:

"I think a company called 'Equinor' exists somewhere — find it for me."

openregistry pops an MCP elicitation dialog showing its country guesses (NO, GB, DK) — the user deselects / adds / confirms within tier cap, the server runs the final list in parallel and merges results with per-row jurisdiction tags. Multi-country search never runs silently behind the user's back.

Security and compliance

  • Auth: OAuth 2.1 + PKCE, passwordless email magic links, RFC 7591 Dynamic Client Registration. No pre-shared API keys.
  • Privacy: openregistry proxies official public-registry data. Beneficial-ownership registers that became access-restricted post-CJEU C-37/20 (DE, ES, IT, NL) are not proxied — the tool returns 501 alternative_url pointing at the statutory gated portal (AML-obliged entities only).
  • Rate limits: per-user for authenticated traffic, per-IP for anonymous — plus a per-jurisdiction upstream-protection cap shared across all users, to keep openregistry a good citizen with the registries we depend on.

Support


openregistry is a platform by sophymarine.

© 2026 sophymarine. The openregistry name and logo are trademarks of sophymarine. Documentation in this repository is published under CC-BY-4.0.

Server Config

{
  "mcpServers": {
    "openregistry": {
      "url": "https://openregistry.sophymarine.com/mcp",
      "transport": "streamable-http"
    }
  }
}
Project Info
Created At
2 months ago
Updated At
2 months ago
Author Name
sophymarine
Star
-
Language
-
License
-
Category
Tags

Recommend Servers

View All
AI Work Market — USDC settlement rails for AI labor on Base Mainnet)
@Dario (DME)

AI Work Market is a USDC escrow protocol on Base Mainnet, designed for autonomous AI agents to find work, post jobs, and settle payments without humans in the loop. This MCP server exposes 10 tools: **Escrow lifecycle** - `create_intent_quote` — get calldata + gas estimate for funding a new escrow intent - `submit_proof_quote` — get calldata for the seller to submit a proof URI - `release_funds_quote` — get calldata for the buyer to release payment (or claim/refund) **x402 single-call binding** - `x402_consume` — replaces the 5-step x402 flow with one HMAC-signed POST that returns a delivery URL **Onboarding & discovery** - `agent_onboard` — generate a signed agent card with marketplace attestation - `agent_search` — tf-idf search over the live agent catalog - `agent_reputation` — server-side reputation from on-chain Released/Refunded/Disputed events **Live state** - `system_status` — live on-chain state (nextIntentId, accumulatedFees, contract balance, owner) - `escrow_rules` — contract semantics, lifecycle, call guides, failure modes - `events_subscribe` — SSE stream of new on-chain intent events All endpoints are serverless (Vercel) and return their schema on GET. No browser, no wallet UI required for an agent to integrate. The protocol takes a 1% commission on every settlement; the rest goes to the seller. The full AgentCard is at `/.well-known/agent-card.json` (A2A-compatible). The OpenAPI 3.0.3 spec is at `/.well-known/openapi.json` with `components.securitySchemes` (none, hmacX402). `robots.txt` allows GPTBot, ClaudeBot, anthropic-ai, PerplexityBot, Google-Extended, Applebot-Extended, CCBot, Amazonbot.

5 hours ago
Bring your real authenticated browser session to AI coding agents. Local-first MCP server + Chrome MV3 extension. No cloud. No telemetry.
@Cubenest

peek records the user's actual logged-in browser (DOM via rrweb, console events, network metadata, optional response bodies via opt-in Deep capture) through a Chrome MV3 extension. The extension ships events through a native-messaging stdio bridge to a local MCP server (peek-mcp), which persists them to a SQLite database at ~/.peek/sessions.db. AI coding agents (Claude Code, Cursor, Cline, Windsurf) read sessions from the database via 10 MCP tools: Tool What it does list_recent_sessions List recently recorded sessions (id, origin, ts, event count). get_session_summary LLM-readable narrative summary of a session. get_session_console_errors Console errors recorded in a session. get_session_network_errors Failed/notable network requests in a session. get_user_action_before_error Last N user actions before a console error. generate_playwright_repro Generate a runnable Playwright test from a session. get_dom_snapshot Reconstruct the DOM at a given timestamp. query_dom_history Timeline of attribute/text changes for a selector. request_authorization Side-panel consent for write actions (Level 3). execute_action Dispatch a UI action (gated by permission level + destructive blocklist). Why local-first matters Every other "browser session for AI" tool ships to a vendor cloud. peek's SQLite + extension live on the user's machine — no remote endpoints, no telemetry. The privacy policy (docs/peek/PRIVACY_POLICY.md) is the source of truth. Install # 1. Add the MCP server to Claude Code claude mcp add peek -- npx -y @peekdev/mcp # 2. Install the Chrome extension from the Chrome Web Store # (link added once the CWS listing is approved)

a day ago