[Agentproto] Re: Sussing out points of disagreement on the charter

["Mirja Kuehlewind (IETF)" <[e-mail]>]

Hi Syed,

> On 12. Aug 2026, at 22:08, Syed Anas Mohiuddin <[e-mail]> wrote:
> 
> Any identity, discovery, or attestation mechanism this group produces must
> state where its trust root comes from, and must not permit that root to be
> derived from the artifact being validated or from data supplied by the party
> being authenticated.

I think we all agree (especially) after the BoF that some kind of identifier is needed. The kind of identifier and mechanism of verification that is required for safe operation needs to be specified in the protocol that is using the identifier and as such is a protocol detail to be discussed later in the group but doesn’t need to be noted explicitly in the charter.

Also note that discovery of agent is not in scope for this potential new group but discussed in dawn. Yes, they might also need some kind of identifier but it easier to make progress of these to steps, agent discovery and agent communication, can be discussed mostly independently. Attestation might be a tool that could be used somewhere but I don’t see it revenant for the agent communication scope, as that really about how to enable agents to manage and schedule tasks. Also attention mechanism itself are better discussion in other groups like RATS.

Mirja