One POST with a website, phone number or email address. Back comes a score from 0 to 10, a verdict, and every signal that produced it — each with its weight. The same engine and the same database as the free checker, with ${DOSSIERS_EN} dossiers behind it.
No black box. Every signal returns with a code, a weight and a sentence you can show a customer.
risk.level | safe · low · medium · high · critical |
risk.score | 0–10, summed from the signals below |
risk.verdict | one sentence, ready to display |
signals[] | per signal: code, weight, explanation — e.g. brandspoof (+6) “possible brand impersonation of ing” |
sources[] | which sources fed this verdict |
checked_at | when the check ran (ISO 8601) |
meta.quota | this month’s usage and limit, in every response |
One POST, synchronous. No SDK, no webhooks, no callbacks — the verdict is in the response.
A customer is sending money to a party they found on a site. Check the domain just before confirmation and show a warning, instead of opening a fraud case afterwards.
Every URL in a listing, chat or DM gets a verdict before another user sees it.
An agent pastes the link or number the customer received and sees immediately whether it is known — with the reasoning for the case file.
If a domain on your platform hits a feed, you see it in the same run instead of after the complaint.
The phone number or email domain given, as one more signal in your existing risk model.
Bearer key in the header, JSON in, JSON out. 60 calls per minute per key; the monthly quota follows your tier.
curl -X POST https://isditveilig.nl/api/v1/check \
-H 'Authorization: Bearer idv_live_…' \
-H 'Content-Type: application/json' \
-d '{"type":"url","value":"https://example.com"}'{
"data": {
"type": "url",
"value": "belast1ngd1enst.nl",
"risk": { "level": "critical", "score": 9, "verdict": "Critical — known scam or abuse." },
"signals": [
{ "code": "brandspoof", "weight": 6, "text": "Possible brand impersonation of 'belastingdienst'" },
{ "code": "no_ssl", "weight": 2, "text": "No valid certificate" },
{ "code": "reach_error","weight": 1, "text": "Site did not respond" }
],
"sources": [ "urlhaus", "spamhaus_dbl", "google_safe_browsing", "internal_safelist", "isditveilig_reports" ],
"checked_at": "2026-09-21T14:30:00+00:00"
},
"meta": {
"rate_limit": { "limit": 60, "remaining": 59, "reset": 1747408800 },
"quota": { "limit": 2500, "used": 1, "period_ends": "2026-10-01T00:00:00+00:00" }
}
}
Documentation →The verdict above runs entirely on our own engine and sources. A few signals we buy in — today line type and carrier for a phone number. Those sit in their own counter per tier, so your invoice stays predictable and so does ours. If your enrichment budget runs out you still get the full verdict, with enrichment.available = false attached. Never an error, never an empty response.
No quote needed to start. Every call tells you how much you have left, and from 80% we warn you — so you never find out after the fact what a month cost.
Anything above your included calls is billed in blocks of 1,000: one call into a new block costs a whole block.
Pick Free or Start, paste the key into your code and you are running. No sales call, no minimum commitment, cancel monthly.
Data processing agreement, SLA, DPIA support, DORA and NIS2 questionnaires, EU-only hosting, SSO and invoice billing. One call, and your security team has what it needs.
Do the sum the other way: what does it cost you today when one customer falls for it — the write-off, the chargeback, the phone call, your service desk's time? Put that next to €129 a month.
Prices exclude VAT, invoiced in EUR. Volume discounts apply above 100k calls/month.
Every verdict cites every signal that produced it. The same sources that power the consumer site — no second-class data for B2B.
Is Dit Veilig? is a free consumer scam-checker. Always has been, always will be. No ads, no affiliate fees, no subscriptions for the people using it.
This API is how we sustain that. Banks, marketplaces, hosting and fraud teams pay for programmatic access to the same data citizens get for free. The same model that funds Wikipedia (free for readers) and Wikimedia Enterprise (paid for Google, Apple, Meta).
See the free check →