Hi Mirja, Henk,

Thanks for the architecture draft and proposed charter. Sharing a use case
from outside the enterprise/workload space that the current scoping doesn’t
fully address.

I’m building Cred (Cred dot ninja), a credential delegation primitive for
this shape of deployment.

Consider a deployment where a consumer authorizes an agent to act on their
behalf across a portfolio of retail sites (Shopify-powered DTC, eBay,
StockX, marketplace platforms, etc.). The retailer doesn’t see the
delegation layer unless they audit signatures.

Two observations:

1. The user-to-operator audit gap is more pronounced for consumer agents
than enterprise ones.

- The bridging artifact (what survives the session and proves the link)
needs to work even when the downstream resource has no OAuth Authorization
Server and the only auth surface is a session cookie.

2. The downstream resource posture varies more than the charter currently
anticipates.

- Agents acting on behalf of users will hit a spectrum of resource auth
shapes. The audit machinery the charter proposes works cleanly for the full
OAuth AS but it needs to compose with non-OAuth resources for the long tail
to be covered.

I’d find it useful if the charter scope included:

- A receipt/evidence format that binds to action payloads independently of
whether the downstream resource speaks OAuth

- Guidance on how delegation chains compose when one or more hops use
non-OAuth auth (cookies, API keys, bespoke session tokens)

Best,

Kieran Sweeney

Principal Engineer
cred.ninja


On Wed, May 20, 2026 at 8:13 AM Mirja Kuehlewind (IETF) <ietf=
[email protected]> wrote:

> Hi wimse group!
>
> We proposed a new potential BoF on auditing (for AI agents mainly) on the
> agent2agent mailing list (see below).
>
> The proposed work could use or would potentially benefit from use of wimse
> identifiers. Therefore we thought this group might be interested in this
> work!
>
> Mirja & Henk
>
> ————————
> Here is also the proposed charter text directly (for more details checkout
> the architecture draft linked below!):
>
> # Agent Use of Delegation and Interaction Traceability (AUDIT) Working
> Group Charter
>
> Autonomous and semi-autonomous software agents, including those based on
> artificial intelligence (AI), are increasingly deployed to act on behalf of
> users, organizations, and services across the Internet. These agents
> interact across multiple administrative or trust domains and can initiate
> actions without direct human oversight at each step.
>
> This introduces challenges for auditability, accountability, and
> transparency, including:
>
> * Difficulty attributing actions to a specific user, agent instance, or
> delegation context
> * Loss of visibility across long-running or distributed workflows
> * Inconsistent capture of delegation relationships, authorization context,
> and identity transitions
> * Cross-domain interactions lack interoperable means to exchange or verify
> audit-relevant information about the participating agents and their
> interactions
>
> AI agents participate in two distinct classes of interactions that must be
> audited:
>
> * User-facing interactions, such as prompts, conversations, and approvals,
> capturing user intent and human-in-the-loop decisions
> * System-facing interactions, such as API calls, tool usage, and
> delegation to other agents or services
>
> Effective auditing requires linking user intent to resulting system
> actions across protocol and administrative boundaries. While traditional
> workflows support evolving authorization, these transitions are usually
> explicit and predefined. AI agent systems introduce dynamic, fine-grained
> authorization changes that arise during execution, driven by agent
> decisions, delegation, and human interaction. Auditing must therefore
> capture authorization as a time-evolving state and correlate these
> transitions across interactions and domains.
>
> Additionally, AI agent behavior may be non-deterministic and not fully
> predefined, requiring auditing mechanisms to capture execution context and
> structure as they emerge. Auditing must also distinguish between user,
> agent, and service identities, and ensure audit data remains interpretable
> across systems without shared assumptions.
>
> ## Scope and Goals
>
> The AUDIT working group will define interoperable mechanisms for auditing
> and accountability of AI agents and delegated systems across Internet
> protocols.
>
> The group will focus on architectures, protocol-layer specifications, and
> data representations that enable systems to record, exchange, and verify
> audit-relevant information across user-facing and system-facing
> interactions. This includes capturing delegation chains, evolving
> authorization state, and enabling consistent interpretation and correlation
> of audit data across domains.
>
> The working group will not define auditing policies or compliance
> frameworks, but instead provide the technical building blocks needed to
> support them.
>
> ## Deliverables
>
> The AUDIT working group is expected to produce:
>
> 1. Architecture for AI Agent Auditing
> An Informational RFC describing roles, trust relationships, and data flows
> for interoperable auditing, including the relationship between user-facing
> and system-facing audit signals.
>
> 2. Audit Data Models and Semantics
> One or more Standards Track RFCs defining data models for representing
> audit information, including interaction records, agent identity,
> delegation context, authorization state over time, and action provenance.
>
> 3. Protocol Extensions or Profiles
> One or more Standards Track RFCs specifying extensions to existing IETF
> protocols (e.g., HTTP, OAuth, or token formats) to convey audit-related
> information.
>
> 4. Best Practices for Deployment and Operation
> An Informational or BCP document providing guidance for secure,
> interoperable, and privacy-aware auditing, including correlation across
> interaction types.
>
> Begin forwarded message:
>
>
> *From: *"Mirja Kuehlewind (IETF)" <[email protected]>
> *Subject: **New draft on AI Agent Auditing and proposed BoF/WG charter*
> *Date: *19. May 2026 at 10:03:46 CEST
> *To: *[email protected]
> *Cc: *Henk Birkholz <[email protected]>
>
> Hi all,
>
> We been working on auditing for AI agents and submitted yesterday a new
> draft that proposes a high-level architecture. Auditing is becoming
> increasingly important for monitoring long-running workflows, trust, and
> also due to regulatory requirements. The proposed architecture takes a
> holistic approach to enable monitoring for the whole agent chain. However,
> please have a look directly at the draft here:
>
> https://www.ietf.org/archive/id/draft-kuehlewind-audit-architecture-00.html
>
> In parallel and given the BoF deadline is already this Friday, we started
> working on a proposed charter. We believe the proposed work would be in
> scope for the IETF and would likely need a new working group, therefore we
> thought it might be useful to provide already a strawman charter to outline
> the potential scope. Please see here:
>
>
> https://github.com/mirjak/draft-audit-architecture/blob/main/audit-charter.md
>
> We would be very interesting to get some quick feedback (before Friday),
> especially about:
>
> 1) Is there general interest in agent auditing work in the IETF? Are there
> other people who maybe already work on this topic or a planning to?
>
> 2) Any comments on the proposed scope of the work? Is the scope clear? Is
> the IETF the right place for this work? Does this need a new working group?
>
> 2) Any comments on the proposed architecture or any additional work in the
> space!
>
> Thanks!
> Mirja & Henk
>
>
> --
> WIMSE mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>
_______________________________________________
OAuth mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to