Trust intelligence API for developers. One call scores an IP, email, phone, domain or device and returns an allow/review/block verdict with every signal behind it. 1,000 lookups a month free, no card, no daily cap.
Screening one signup used to mean calling three vendors: one for IP reputation, another for email, a third for phone. Three keys, three quotas, three invoices, and the logic tying them together was yours to write.
So we built the thing we wanted. One call takes an IP, email, phone, domain or device — or a whole signup at once — and returns an allow/review/block verdict with every signal that produced it.
The signals matter more than the score. "Our vendor said 87" is not an answer you can give a customer, and it gives you nothing to tune. So every response carries its reasons, and strictness is yours to set.
Our stack
Next.js on Vercel Fluid Compute, Postgres on Supabase, DB-IP for geolocation, Dodo Payments as merchant of record.
Node/TypeScript and Python SDKs, a CLI, Express and Next.js middleware, and an MCP server so coding agents can call the API directly. The SDKs ship with zero dependencies, on purpose — a fraud check should not drag a tree of transitive packages into your build.