Every fraud tool on the market — IPQS, Sift, DataDome — was built for a world where bots used datacenter IPs. That world is gone.
Today, fraudsters use antidetect browsers like Kameleo, GoLogin, AdsPower, and Dolphin to spoof their fingerprints. They route traffic through residential proxies that look like normal home connections. Legacy tools can't see them.
So I built Sentinel.
What It Does
Sentinel is a REST API that tells you if a visitor is real or fake. One API call, under 40ms, checks:
- VPN / Proxy detection (including residential proxies)
- Antidetect browser detection (Kameleo, GoLogin, AdsPower, Dolphin)
- AI bot detection (LLM-powered scrapers that mimic humans)
- Device intelligence (incognito, VM, emulator, browser tampering)
- 3D face liveness (optional — proves the user is a real person at the camera)
How It Works
1. Add the SDK to your frontend:
html
<script async src="https://sntlhq.com/assets/edge.js"></script>
2. Call the API from your backend:
const result = await fetch('https://sntlhq.com/v1/evaluate', {
method: 'POST',
headers: {
'Authorization': 'Bearer sk_live_YOUR_KEY',
'Content-Type': 'application/json'
},
body: JSON.stringify({ token: req.body.sentinelToken })
});
const data = await result.json();
// data.isSuspicious → true = block, false = allow
3. Get a clear answer:
{
"isSuspicious": false,
"details": {
"ip": "84.15.201.92",
"vpn": false,
"proxied": false,
"cc": "EE"
},
"deviceIntel": {
"browserTampering": false,
"botDetected": false,
"incognito": false
}
}
Why I Built This
I worked in fraud prevention and kept seeing the same problem: tools that cost $500+/month and still missed 40% of modern threats. Residential proxies and antidetect browsers make traditional IP reputation useless.
Sentinel combines network intelligence (IP, VPN, proxy detection) with device intelligence (fingerprint tampering, bot detection) in a single call. No CAPTCHAs, no friction for real users.
Free to Use
1,000 API requests per hour — no credit card required
Sign up at sntlhq.com and get your API key in 2 minutes
Full docs at sntlhq.com/api
Would love feedback from anyone working in fraud prevention or security. What signals would make this more useful for your use case?
Top comments (0)