Verifiable credential
A tamper-evident, cryptographically signed claim about a subject that a third party can verify without contacting the issuer — the digital equivalent of a signed certificate.
A verifiable credential is a digitally signed statement about a subject — for example, that an agent belongs to a particular organization or is authorized for a task. Because it's signed by an issuer and bound to the subject's identifier, anyone can verify it without phoning the issuer.
Verifiable credentials are a W3C standard and pair naturally with DIDs: the credential references the subject's DID, and the holder presents it along with proof they control that DID.
Issuer, holder, verifier
Three roles define the model. An issuer signs a credential about a subject. A holder — often the subject — stores it and presents it when asked. A verifier checks the issuer's signature and that the holder controls the referenced identity.
The verifier trusts the credential based on the issuer's signature, not on a live lookup, which makes the check fast and offline-friendly.
Where this is heading for agents
As agents act on behalf of people and organizations, verifiable credentials let them prove authorization — that this agent is allowed to do this thing — in a way a counterparty can check independently. It builds on the same identity foundation as did:web and Ed25519 signing.
FAQ
How is this different from an API key?
An API key is a shared secret for one service. A verifiable credential is a signed claim any verifier can check, tied to the subject's identity.
Does verification need the issuer online?
No. The verifier checks the issuer's signature, so the check works without contacting the issuer in real time.
Give your agent an inbox.
A real email address, a vault, 2FA, and an identity in one API call.