Clarify the human-authorization model after did:wba v1.1

[Math1987]

Hi — I reviewed the current ANP specifications and AgentConnect before proposing any change, and I would like to clarify the intended human-authorization model.

Issue #14 documented the earlier approach in which humanAuthorization was a dedicated DID verification relationship/key.

The current documents now appear to describe a different layering:

• ANP-03 v1.1 says that DID Documents no longer define humanAuthorization. Human confirmation belongs to an upper-layer authorization policy, and the document gives authorizationLevel: user-presence-required as an example.
• ANP-07 defines the interface-level boolean humanAuthorization.
• Draft ANP-06 uses requiresHumanAuthorization as a capability or negotiation constraint and explicitly states that it does not mean authorization has been completed.
• The technical white paper still describes humanAuthorization as a dedicated DID verification method/key that must sign high-risk requests.

Could you confirm whether the intended model is:

1. did:wba authentication proves the caller/request identity;
2. ANP-07 humanAuthorization declares that an interface requires human approval;
3. ANP-06 requiresHumanAuthorization expresses that requirement during capability discovery or negotiation;
4. completion and verification of the actual approval are left to the upper-layer business protocol and local authorization policy?

It would also be helpful to clarify whether authorizationLevel in ANP-03 is only an illustrative example, or whether it is intended to become an Agent Description field alongside or instead of humanAuthorization.

If this interpretation is correct, would a small documentation-only PR be welcome to:

• align the technical white paper with ANP-03 v1.1;
• add a short cross-reference between ANP-03, ANP-07, and ANP-06;
• explain the distinction between declaration, negotiation, and actual approval;
• update the relevant English and Chinese documentation together, so that the terminology and authorization model remain clear and consistent across both versions?

I would keep this purely editorial, without proposing a wire-format change.