"Verified" means nothing unless you know what was checked. Credia Verify checks specific claims about an agent's operator, such as their domain, GitHub account or organisation, and records how and when each check was done and when it expires.
Available to selected teams. Request access to discuss your integration.
Credia Verify
Status: Private Beta
Inspect specific claims and the evidence behind them.
You want the claims about your agent, such as who operates it, to be checked rather than only declared.
Agents call your app, API or marketplace, and you need to know what has actually been confirmed about them before you trust them.
Many platforms show a "verified" badge without saying what it means. Was an email confirmed? A company checked? Something else? And is it still true today?
Credia Verify checks one specific claim at a time, for example that an agent's operator controls a certain domain. Each result records what was checked, how, when, and when it expires. If the claim stops being true, the result changes and the badge goes away.
For example domain, GitHub, email or organisation, for an agent's operator.
Add a DNS record, sign in with GitHub or confirm an email, depending on the check.
What was checked, how, when, the result and the expiry date.
Checks re-run on schedule. If one fails, the result and badge update automatically.
Situations where knowing exactly what has been checked matters.
Your marketplace lists AI agents. Anyone can sign up, and you want a basic bar before an agent appears.
ఇది ఎలా పనిచేస్తుంది
Every listed agent has a checked operator domain, and the bar stays enforced over time.
An agent shows a "verified" badge. You want to know what that actually means before you rely on it.
ఇది ఎలా పనిచేస్తుంది
You trust what has been checked, and only that.
You verified your domain last year, then moved to a new one. An old badge that is no longer true hurts your credibility.
ఇది ఎలా పనిచేస్తుంది
Your agent never shows a badge that has stopped being true.
A partner or auditor asks how you knew who operated an agent at the time you worked with it.
ఇది ఎలా పనిచేస్తుంది
A clear answer about what you knew, and when you knew it.
The same lifecycle for every type of check.
For a specific claim, such as control of a domain.
For example by adding a DNS record or signing in with GitHub.
Passed or failed, with the method, time and expiry date.
On the agent card, and to platforms over the API.
On schedule. A failed re-check updates the result straight away.
An illustrative example of a check result. The exact beta contract is shared with teams during onboarding.
{
"agentId": "agt_7Hq2k9",
"claim": { "type": "domain_control", "value": "acme.example" },
"method": "dns_txt_record",
"status": "passed",
"checkedAt": "2026-11-02T09:14:00Z",
"expiresAt": "2027-02-02T09:14:00Z",
"nextRecheck": "2026-12-02"
}Credia Verify builds on the email, GitHub and domain checks shown on AgentCredia agent pages today. A paid plan may unlock a check; payment itself never awards a result.
It is not KYC or KYB: it does not check passports or company registration documents. It does not rate an agent as safe, or certify its reasoning, quality or future behaviour. Each check means exactly what it says, and nothing more.
That the operator controlled the domain when the check ran. It does not prove legal ownership of a business.
No. Credia Verify checks specific claims such as domain, GitHub and organisation control. Identity documents and company registrations are not part of the beta.
Things change: domains lapse and accounts move. Expiry and automatic re-checks keep results current instead of showing a badge that stopped being true.
No. A badge states one checked claim. It is evidence for your decision, not a safety rating.
Credia Verify is in Private Beta. Request access and tell us which checks your workflow needs.
Establish an agent's identity and its connection to an operator.
Explore IdentityDefine how an agent authenticates and what authority it can use.
Explore AuthUnderstand activity and outcomes in context over time.
Explore ReputationPayment workflows and controls for AI agents.
Explore PaymentsSee if the traffic in your app is human, AI-assisted or agent traffic.
Explore Proof-of-ActorTell us which claims your workflow needs to check and who relies on the result.