A live person check for AI agents, APIs, and products — anywhere a workflow needs to confirm a real human before releasing payouts, resetting accounts, retrieving sensitive data, or approving claims.
Same DocVerify forensic posture, now where your agent or API needs to confirm an actual human is on the other end. Stop the workflow, verify the person, then continue.
0.4%
false-pass rate against replayed selfie attacks in internal benchmarks.
if (action.risk === 'high') {
await docverify.realPersonCheck({
user_id,
reason: 'release_payout',
})
}Pause an agent or API workflow until the user proves they are a live, real person on camera.
Guided movement and camera quality signals separate a live human from replayed or generated media.
Pair the live capture with passport or ID document evidence and return it as structured context.
Continue, block, retry capture, or escalate to a human with the full evidence trail attached.
Workflow
An agent, API, or product flow touches money, private data, regulated access, or account ownership.
The user completes hosted liveness capture and ID document evidence in a secure browser flow.
Liveness, document, motion, replay, and quality signals come back as structured JSON.
Your code continues only when the evidence proves a real person and satisfies your policy.
curl -X POST \
-H "X-API-Key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"type":"real_person_check","reason":"release_payout"}' \
https://docverify.app/api/id-verifications{
"session_id": "rpc_9x24...",
"reason": "release_payout",
"status": "verified",
"signals": {
"document_capture": "passed",
"liveness_motion": "passed",
"replay_risk": "low"
}
}Your agent or API should not rely on chat text or a session cookie when the next action matters. Start hosted verification, collect live human evidence, and return a structured result your code can branch on.
Use it from agents, APIs, products, CRM, support flows, or compliance queues.
Designed for sensitive identity evidence and controlled review access.
Answers below are mirrored in structured data for search engines.
A real person check confirms there is a live human on the other end of an agent, API, or product workflow before it performs a sensitive action. It pairs guided liveness motion with ID document capture such as a passport or ID card.
Anywhere a workflow needs proof of a real human: AI agents about to release payouts, reset accounts, retrieve sensitive data, approve claims, or escalate; APIs that gate high-risk actions; and products that need to confirm an account owner is actually present.
It helps reduce static selfie spoofing, screen replay, printed photo attacks, AI-generated face attacks, synthetic onboarding attempts, and weak evidence that an agent or API should not trust automatically.
Yes. Your product, agent backend, or API starts a verification session, sends the user through hosted capture, then receives a structured result you can use in policy checks, tool calls, and workflow gating logic.
Let your agents, APIs, and products verify a live human before they approve, retrieve, transfer, reset, or escalate.
This site uses cookies for authentication and analytics. Free-tier uploads may be retained to improve our models; paid-tier uploads are never stored. Learn more