Documentation

Worldwide drug data, three ways in.

The same catalogue backs all three: a console you can search, a REST API you can integrate, and an MCP server your AI client can call. Pick the door that matches what you came to do — nothing is hidden behind the other ones.

Common questions

Three things everyone asks.

  1. 01

    Where does the data come from, and how fresh is it?

    Public national registers and open scientific databases only — Swissmedic, BAG, the EMA, national medicines agencies across the EU, and the FDA. Every response carries a _meta.sources block stamping each source with when it last synced successfully.

    Sources and freshness
  2. 02

    Can I be told when something changes instead of polling?

    Yes. Subscribe a URL to the change feed and we POST a batch with an HMAC-SHA256 signature whenever a register moves. Requires the webhooks:write scope, which starts at the Starter plan.

    Webhooks
  3. 03

    Do you process patient data?

    No. The prescription-screening endpoint accepts drug identifiers and coarse, non-identifying context only — age band, sex, pregnancy flag, eGFR band — and rejects any request carrying a patient identifier with a 422 rather than scrubbing it silently. Nothing patient-shaped is stored or forwarded.

    The full answer in the FAQ

Reference

The words, and the answers.

Reading this signed out? Every page here is public. When you want to run a call against your own key, create an account — the Free plan needs no card — or open the interactive explorer and paste a key you already have.