You are here. See how this question connects to other ideas.
Select a node to open its page · Expand to read within the map
One question
Are agent identity and permission the same?
Separate persistent naming, proof of control and authority for an action.
For cross-application action, distinguish the identifier, its controller, the represented party and the permitted action.
A DID can identify software. Verification methods establish key-control relationships; credentials can carry issuer claims. Verifiers still evaluate issuer trust, credential status and business authorization. A DID grants neither reputation nor permission by itself.
Why connect this to blockchain?
Some DID methods use a chain; others do not. Identifiable actors and enforceable authority do not require every agent to have an on-chain account.
A purchasing agent might receive task-scoped authority instead of the owner’s root key. Scope, budget, expiry and revocation need separate design. This is an architectural example, not a claim that a complete delegation product has shipped.
Sources and further reading
Check your understanding
Can an agent that signs a message automatically pay for a company?
No. Identifier control does not establish company payment authority.
Continue along a learning path
- From AI to agents that actStep 3 of 5