# WhiteIntel — Ownership Intelligence

> UBO, sanctions & ownership graph. 23 tools, 130.7M entities, 31 registries.

- name: dev.whiteintel/whiteintel
- version: 0.7.10
- connection class: unknown (unknown)
- trust: 63/100
- quality: 81/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/dev.whiteintel%2Fwhiteintel
- tools: https://api.protogrid.dev/v1/servers/dev.whiteintel%2Fwhiteintel/tools

**Not classified yet.** The remote endpoints have not been probed successfully, or the listing has nothing to connect to.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "whiteintel": {
      "type": "http",
      "url": "https://whiteintel.dev/api/mcp"
    }
  }
}
```


## Trust

- hygiene: 100
- liveness: 0
- freshness: 85
- provenance: 85
- drivers: +repository +dns-namespace +website -unreachable(12 failures) ~updated-61d-ago
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 81/100 (good)

### protocol: 75 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-11-25; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- pass `protocol.transport`: streamable HTTP
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 93 (weight 30)

- pass `tools.descriptions`: every tool has a description
- warn `tools.description_length`: 2 tools have descriptions over 1,024 characters, which crowds the context
- pass `tools.schemas`: every tool has an object input schema
- pass `tools.annotations`: 24 of 24 tools declare readOnlyHint or destructiveHint
- pass `tools.directory_hints`: every tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHint
- pass `tools.token_cost`: about 3,792 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: n/a (weight 15)

- n/a `stability.changes`, `stability.rug_pull`: tool history not recorded yet

### dependencies: 67 (weight 15)

- pass `deps.known_vulns`: no known advisory in 98 resolved packages
- fail `deps.mcp_sdk_version`: @modelcontextprotocol/sdk 1.29.0 has 1 known advisory (GHSA-6qxp-vccf-f47h); update @modelcontextprotocol/sdk to 1.31.0 or later
- pass `deps.resolvable`: @whiteintel/mcp-server@0.7.10 resolved: 98 packages

- tools: 24, about 3,792 tokens to load them all
- badge: [![protogrid quality](https://protogrid.dev/badge/dev.whiteintel/whiteintel.svg)](https://protogrid.dev/servers/dev.whiteintel/whiteintel)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

_No tool definition changes recorded._

## Dependencies

- npm @whiteintel/mcp-server@0.7.10: resolved, 97 dependencies, MCP SDK @modelcontextprotocol/sdk 1.29.0
  - GHSA-6qxp-vccf-f47h (high) in `` @modelcontextprotocol/sdk@1.29.0 ``, direct, fixed in `` 1.31.0 ``, advisory summary: "MCP TypeScript SDK: OAuth client could send credentials to an authorization server chosen by the MCP server" https://osv.dev/vulnerability/GHSA-6qxp-vccf-f47h

Dependency graphs from deps.dev; advisories from OSV.dev, including the GitHub Advisory Database and the PyPI Advisory Database; all CC BY 4.0.

## Remotes

- https://whiteintel.dev/api/mcp (streamable-http, auth none, reachable false, uptime30d 0.47826087)

## Packages

- npm @whiteintel/mcp-server@0.7.10 secrets: WHITEINTEL_API_KEY

## Tools (24)

- `buy_dossier`: Start a one-off dossier purchase via guest Stripe Checkout — no WhiteIntel account needed (Stripe collects an email for delivery). Pick a tier ('standard' €39: full UBO chain + financial history · 'premium' €99: additionally itemised assets) and optionally a bulk pack ('5'/'25'), plus the entity_id the report is for. Returns checkout_url + next_steps: open the URL to pay, then feed the session_id to claim_dossier. See get_pricing for the full list.
- `check_offshore_exposure`: Walk the ownership chain and flag sanctioned + secrecy-jurisdiction hops — the offshore-layering lead.
- `claim_dossier`: Redeem a paid Stripe Checkout session for a dossier access token. Pass the session_id (cs_…) from the post-payment redirect after buy_dossier. Returns { token, entity_id, tier } — pass the token to get_dossier as its `token` input. Idempotent; fails with 402 not_paid until payment completes, so wait for the human to finish Checkout then call again.
- `find_similar`: Entities most similar to a given one — the nearest corpus dossier cards ('more like this') for peer discovery. Pass an entity id from search_entities / semantic_search. COVERAGE IS PARTIAL — only entities in the embedded risk-scored subset (~1.9% of the corpus and growing) return peers; an entity outside it returns empty for now, not an error.
- `get_company_details`: Registered profile for a company: address, status, SIC, incorporation, plus filing/compliance (overdue, charges, former names).
- `get_dossier`: Structured, fully-cited dossier for one entity: cross-source identity, ownership/UBO chain, risk signals, filed financials, provenance. Pass the `token` from claim_dossier to unlock the full paid depth you purchased for this entity.
- `get_entity`: Full record for one entity id + its direct relationships, with provenance.
- `get_financials`: Filed financials year-over-year: turnover, profit, net assets, cash, employees (Companies House iXBRL).
- `get_payment_link`: PERMANENT, shareable Stripe payment links for the one-off dossiers — use this instead of buy_dossier when you need something you can HAND TO A HUMAN. buy_dossier mints a cs_live_ Checkout Session that is single-use and expires in 24h, so it is useless in a report or a message the human reads tomorrow; these links never expire and can be reused. Append ?client_reference_id=<entity uuid from search_entities> to bind the purchase to one company — without it the buyer gets a dossier credit, spendable on any entity later. No API key needed, no WhiteIntel account needed.
- `get_pricing`: WhiteIntel's price list + the exact machine flow for buying access: one verified dossier (€29, one price, everything), subscriptions (Pro €99/mo, Scale €499/mo) and the metered API. Returns how_an_agent_buys — buy_dossier opens a Stripe Checkout, claim_dossier mints the token, get_dossier with that token returns the unlocked report. Static, no network — check it before recommending a purchase.
- `get_pulse`: The corpus activity feed — recent ownership-change and filed-accounts events, cited. Pass since=<next_since> to stream only new events. The default (unfiltered) feed returns only events that carry a source URL. `watchlist` (OpenSanctions PEP listings) is opt-in via kind=watchlist and is currently uncited (source-url NULL for every row).
- `get_sanctions`: An entity's sanctions exposure (OFAC/EU/UN/UK) for it and its resolved cluster siblings, each cited.
- `graph_neighbourhood`: Every ownership/control edge within a bounded number of hops of one entity, in BOTH directions — who it controls, who controls it, and their neighbours. Hard-capped in the database: depth 3, 300 edges, and at most 25 edges followed per entity per direction per hop. When the edge budget runs out the response sets `truncated: true` and says so — the corpus contains single entities with more than 22,000 edges, so a truncated view is normal for hubs, not an error.
- `graph_path`: How two entities are connected: the ordered hops of a bounded breadth-first search over ownership and control edges in both directions. ⚠️ BOUNDED, NOT EXHAUSTIVE — at most 15 edges are followed per entity, per direction, per hop, so `found: false` means NO PATH WAS FOUND WITHIN THOSE BOUNDS and is NOT evidence that the two entities are unconnected. The response always carries `exhaustive: false`; never report a negative result as a clean bill of health.
- `list_asset_coverage`: The WhiteIntel ASSET-ownership coverage map — who owns the plane / yacht / property, and HOW we hold that link. The companion to list_jurisdictions for physical assets. Each row is one asset class × area with a `tier`: `deep` = we ingest a bulk source that ties the asset to an OWNER (e.g. FAA US aircraft, HM Land Registry UK property) · `indexed` = held only via leaks/sanctions (e.g. a yacht reached through an offshore SPV in the ICIJ leaks) · `on_demand` = the source is closed/paid, so the record is procured from source on a paid request · `community` = SnitchBoard crowd tips. Also `links_to_owner` — CRITICAL, because many asset registries publish only the registration mark and NOT the owner: an offshore aircraft register (Isle of Man, Bermuda) names the SPV/owner-trust, not the human behind it, so `deep` there is still the SPV layer, and piercing to the beneficial owner is an `on_demand` bizjet-ownership buy. Read this before claiming we do or do not hold ownership for an aircraft/vessel/real-estate entity. No vendor or price is exposed. Returns { assets, count, classes, tiers, note }.
- `list_jurisdictions`: The WhiteIntel coverage map — every jurisdiction we hold and HOW we hold it. Read this before telling anyone a country is or is not covered, because 'covered' means three different things. Each row carries `tier`: `deep` = we loaded the country's WHOLE national registry, so a name/number search resolves ANY company registered there; `indexed` = we hold only the leak / sanctions / GLEIF subset, so the entities that surfaced in a leak or on a sanctions list are searchable but the rest of that country's companies are NOT in the corpus; `on_demand` = the registry is closed or paid, so the specific record is procured from source when a dossier is purchased. Also `scope` (full = whole registry · subset = fragment), `depth` (`ownership` = owners/beneficial owners on the record · `officers` = directors · `identity` = name/number/address/status, owners procured on request) and `registry` (our loader, for deep tiers). So a `deep`+`full`+`ownership` row (e.g. gb, lv, ua, br) means you can trace owners for any company there; a `subset` row (e.g. cn, kr, most secrecy havens) means an empty search is 'not in the held subset', NOT 'does not exist' — the full record is bought on request. No per-record price or vendor is exposed. Returns { jurisdictions, count, tiers, note }.
- `lookup_by_identifier`: Resolve a corpus entity by a strong identifier — lei | ofac | eu | un | uk | uen | sec | krs | gb-coh | siren | br-cnpj (Brazil RFB CNPJ, 8-digit root or full 14-digit).
- `lookup_company`: UK company by Companies House number → record + ready-built ownership graph.
- `purchase_dossier`: BUY the full dossier for one entity programmatically — no browser, no Stripe Checkout. Debits your prepaid wallet €2 and returns the complete premium dossier (full UBO chain, filed financials, itemised assets, provenance) in the SAME response. Requires a funded API key (Authorization: Bearer wi_…). Idempotent per (buyer, entity): buying the same entity again returns it with NO second charge, so a retry is safe. If the wallet balance is too low the call returns HTTP 402 with a machine-readable payment requirement + a top-up URL — hand the top-up to a human once, then retry; the agent spends from the balance thereafter with no browser in the loop. Get the id from search_entities / resolve. This is the AGENT-NATIVE purchase path; buy_dossier is the human/browser (Stripe Checkout) path.
- `resolve`: Batch-resolve a list of company names or scheme:value identifiers (lei/siren/br-cnpj/gb-coh/uen/sec/ofac/eu/un/uk/krs) to canonical WhiteIntel entity ids + confidence in ONE call. Enrich a whole supplier/counterparty list without one lookup per row. Up to 25 anon / 100 keyed.
- `search_companies`: Free-text UK Companies House company-name search → registration number.
- `search_entities`: Search every node in the corpus — companies AND people — by name. Returns entity ids for get_dossier / trace_ownership_path.
- `semantic_search`: Meaning-based entity search (BGE-M3 vector ANN over the resolved dossier cards). Finds companies/people whose profile is semantically closest to a natural-language query even without a keyword match. Optional kind + jurisdiction filters. COVERAGE IS PARTIAL — the risk-scored subset of the corpus is embedded so far (~1.9% and growing with the backfill); a thin or empty result is NOT proof the entity is unknown, so pair with search_entities (lexical/name) before concluding an entity does not exist.
- `trace_ownership_path`: Walk ownership upward from a root entity to the ultimate beneficial owner(s).