> ## Documentation Index > Fetch the complete documentation index at: https://docs.shieldlabs.ai/llms.txt > Use this file to discover all available pages before exploring further. # Fraud detection and prevention docs > Detect risky users, devices and IPs under any masking, and stop multi-accounting, account sharing and takeovers.

Detect risky users
under any masking

Stop multi-accounting, account sharing and takeovers. ShieldLabs scores every user, device, visitor and IP address, with 99.9% identification accuracy and 99.9% risk signal detection accuracy. Integration takes five minutes.

Get started Explore the API

Five-minute integration · API and webhooks

User
User HIDe3b0…b924
RiskDANGEROUS
Linked devices2
Linked public IPs4
High-Risk Event
Multi-accountingHIGH CONFIDENCE
Latest identification
Device IDd290…0851
VPN+15
Anti-detect Browser+60
Risk Score75DANGEROUS

Detect risky users. Stop abuse of your product.

One integration gives you your users, devices, visitors and IPs with their risk, the four High-Risk Events, every risk signal and your traffic quality.

Multi-accounting, Account sharing, Impossible travel and Account takeover, detected on your users out of the box, each at Medium or High confidence.
Multi-accountingAccount sharingImpossible travelAccount takeover
 Link every user to the devices, visitors and IP addresses they use, each with its own risk band. The Device ID holds through cleared cookies, incognito mode and IP changes.
User HIDDevice IDVisitor IDPublic IPLocal IP
 Catch the masking and automation behind risky users, well beyond IP blocklists: VPNs, proxies, Tor, datacenter IPs, anti-detect browsers, bots and browser automation.
VPNProxyTorAnti-detectBots+ more
 Every identification gets a Risk Score from 0 to 100 with each risk signal named and weighted. Users, devices, visitors and IPs carry the worst band of their identifications.
0-100TrustedSuspiciousDangerous
 Score your traffic quality by channel, referrer and UTM campaign.
ChannelsReferrersUTMTraffic quality
 The Overview screen of the analytics dashboard for the last 7 days: traffic quality at 14.36, Trusted, over 12,480 identifications; 1,240 users with 104 risky users and 46 High-Risk Event users; the four High-Risk Events with their Medium and High confidence counts; 3,910 unique visitors, 3,490 Trusted and 420 risky; the top risk signals VPN, Timezone Mismatch, Datacenter IP and Proxy; and the riskiest identifications. The Overview screen of the analytics dashboard in the dark theme for the last 7 days: traffic quality at 14.36, Trusted, over 12,480 identifications; 1,240 users with 104 risky users and 46 High-Risk Event users; the four High-Risk Events with their Medium and High confidence counts; 3,910 unique visitors, 3,490 Trusted and 420 risky; the top risk signals VPN, Timezone Mismatch, Datacenter IP and Proxy; and the riskiest identifications.

Tour the analytics dashboard

Start by use case

Each recipe is a worked, copy-pasteable flow built on your users, their devices and IPs, and the Risk Score of each identification.
View all use cases

Protect your users
Account takeover is detected on the user at Medium or High confidence. Step up before a sensitive action. Tell a spray of scripted logins apart from your real returning users. Skip friction for known devices, step up only when a login looks risky.
Grow your revenue
Catch one device or local IP opening many fresh accounts at signup. Stop multi-accounting from draining coupons and signup bonuses. Detect one account used from several distinct devices, at Medium or High confidence. Recognize the same device through cleared cookies and incognito mode. Catch VPNs and proxies used to reach a cheaper region's price. Recognize a trusted device on return and spare it repeated checks.
Protect your platform
Keep a banned user's device out, even when they come back with a fresh account. Attach the account's device and identification evidence to a disputed order. Rank affiliate partners by the risky accounts and self-referred traffic they bring. Grade each channel by the users and devices it brings and its traffic quality. Add friction at payment only when the buyer's account or identification looks risky.

Built for easy integration

The JavaScript snippet, the Server API and webhooks. ShieldLabs returns a Risk Score with every named risk signal on each identification. You choose the action for each case and act on it in your backend.

See the full API 
          {`const mod = await import(
                    'https://cdn.shieldlabs.ai/snippet.js?publicKey=YOUR_PUBLIC_KEY'
                    );
                    // Signed-in user: pass a hashed account id
                    mod.checkAuthenticatedUser('HASHED_USER_ID');
                    // Page without a signed-in user:
                    // mod.checkAnonymous();`}
  • Identify users and devices with 99.9% identification accuracy
  • Works with React, Vue, Svelte, Angular, and plain JS
  • One ES module from cdn.shieldlabs.ai, loaded with a dynamic import
Read the snippet docs › 
          {`curl "https://account.shieldlabs.ai/api/v1/history/user_hid/HASHED_USER_ID?limit=20" \\
                    -H "Authorization: Bearer sec_your_private_api_key"`}
  • Read every identification of one user, device, visitor or IP
  • Authenticate with the Private API Key from Integration > API keys in the analytics dashboard
  • Guaranteed read by request ID when a webhook is missed
Read the Server API docs › 
          {`POST /your/webhook
                    X-Shield-Signature: sha256=

                    {
                    "event_type": "identification.scored",
                    "schema_version": "2026-06-01",
                    "created_at": "2026-06-16T18:00:21Z",
                    "data": {
                      "request_id": "550e8400-e29b-41d4-a716-446655440000",
                      "user_hid": "e3b0c442f8a1b924",
                      "risk_score": 70,
                      "signals": [
                        { "name": "antidetect_browser", "weight": 60 },
                        { "name": "proxy", "weight": 10 }
                      ]
                    }
                    }`}
  • Pushed to your backend about 300 milliseconds after the check
  • Verify X-Shield-Signature on the raw body
  • At-most-once, so make your handler idempotent on request\_id
Read the webhook docs ›

Keep going

Install the snippet, identify your first user and read a Risk Score in five minutes. No webhook, no risk signals, or a Risk Score you did not expect? Start here. What shipped recently across the snippet, the Risk Score and the analytics dashboard.
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.