AbuseGraph
See the graph
behind every signup.
Browser probes plus an identity graph that links devices, emails, and prior exposure — so fake accounts don't look like real users.
Free tier · Hosted API · No credit card
What AbuseGraph does
Four layers. One identity graph.
Signup abuse shows up as patterns across sessions. We connect those patterns — then hand you a score you can act on.
Identity graph
Link visitors by device, email, IP, browser signature, and prior exposure. Catch multi-account abuse even when they switch networks.
- Cross-account hops across sessions
- Shared device and email clustering
- Flags returning blocked identities
Device signals
Client-side probes feed the graph — automation markers, environment consistency, and extension cues with zero telemetry.
- Works across sessions and incognito
- Local computation only
- Optional SDK for richer probes
Clear verdicts
Every check returns a 0–100 score and a verdict: allow, challenge, shadow, or block. You enforce — we never touch passwords.
- Transparent signal breakdown
- Four actionable verdicts
- Hosted /api/v1/check API
Bot & agent detection
Catch headless browsers, anti-detect tools, and agent-style extensions without putting CAPTCHAs in front of real users.
- Headless and automation markers
- Anti-detect lie detection
- Suspicious extension scanning
How it works
Three steps from install to protection.
Optional browser SDK gathers device probes. Your backend sends email + IP to AbuseGraph.
We link the event into the identity graph — devices, emails, velocity, and prior exposure.
You get a score, verdict, and plain-English reasons. Enforce allow / challenge / block yourself.
No CAPTCHA tax
Real users never see a challenge from AbuseGraph. Bots are scored via probes and graph links.
62 risk signals
Automation markers, environment consistency, extensions, TLS, and identity-graph linkage — each visible in Review.