How to verify any AI agent credential in under 10 lines of code

#ai #agents #security #tutorial

The 10-line verification

import requests

def verify_credential(credential):
    r = requests.post(
        "https://www.marketnow.site/api/trust?action=verify",
        json={"payload": credential}
    )
    return r.json()

# That's it. No API key. No auth. No SDK.
result = verify_credential({
    "@context": ["https://www.w3.org/2018/credentials/v1"],
    "type": ["VerifiableCredential"],
    "issuer": "did:web:alice.example",
    "credentialSubject": {"id": "did:agent:bob"},
    "proof": {"type": "Ed25519Signature2020", "verificationMethod": "did:web:alice.example#key-1", "proofValue": "placeholder"}
})

print(result["valid"])  # True
print(result["format"])  # w3c-vc
print(result["uts"]["trust"]["score"])  # 5

What you get back

The API returns:

valid — true/false
format — which of the 8 formats was detected
uts — the Universal Trust Schema with trust score, confidence, subject, revocation status
issues — list of problems if invalid
warnings — non-blocking warnings

8 formats auto-detected

You don't need to tell the API what format your credential is. It auto-detects:

ATC v3 (Agent Trust Card)
JWT (OAuth/OIDC)
W3C Verifiable Credentials
A2A (Google Agent-to-Agent)
EAT-AI (IETF Entity Attestation)
ZTA (Zero Trust Agent)
MCP Server Cards
X.509 certificates

In JavaScript

const result = await fetch("https://www.marketnow.site/api/trust?action=verify", {
    method: "POST",
    headers: {"Content-Type": "application/json"},
    body: JSON.stringify({payload: credential})
}).then(r => r.json());

if (!result.valid) {
    throw new Error(`Verification failed: ${result.issues}`);
}

In Bash

curl -X POST "https://www.marketnow.site/api/trust?action=verify" -H "Content-Type: application/json" -d '{"payload": "..."}'

In any language

The API is a simple HTTP POST. No SDK needed. No API key. No registration. CORS is open.

DEV Community