Re: Agent-Native Auth and Id: auth.md as a Signal of Emerging Standards Gaps

Thank you Aaron, yes.

Is the overview below describing a possible divergence in the landscape
right?
Let us know what you make f it thanks PDM

Gap Analysis

The emerging AI-agent identity stack is converging around two historically
separate standards ecosystems: the IETF/OAuth authorization model and the
W3C verifiable credential/decentralized identity model. Projects such as
auth.md illustrate this convergence clearly. auth.md is architecturally
aligned with OAuth, HTTP metadata discovery, and IETF identity assertions
rather than W3C DID/VC specifications, yet it is attempting to solve
adjacent problems: portable machine identity, delegated authority, trust
propagation, and interoperable agent registration. This creates a growing
standards gap. The current IETF stack is highly effective for
authentication, authorization, delegated access, enterprise federation, and
API-native workflows, but it is weaker at portable attestations, long-lived
trust portability, and cross-domain machine reputation. Conversely, the W3C
VC/DID ecosystem is strong at credential portability and cryptographic
attestations but comparatively weak in operational enterprise deployment,
authorization semantics, and large-scale service interoperability.

The practical implication is that AI agents increasingly require both
models simultaneously. An autonomous agent may need to authenticate to APIs
via OAuth, carry signed organizational credentials, prove delegated
authority, expose machine-readable metadata, maintain portable reputation,
and support cryptographically verifiable provenance across domains. No
single standards family currently provides a complete solution. As a
result, the most strategically important area is not competition between
the IETF and W3C ecosystems, but interoperability between them. The likely
direction of travel is a hybrid architecture in which OAuth/OIDC and
emerging IETF identity drafts govern operational access control, while
verifiable credentials and selective disclosure mechanisms govern portable
assertions and trust claims. The commercial market already appears to be
converging around pragmatic interoperability rather than ideological “pure”
decentralization or fully centralized platform identity.
Functional Area IETF / OAuth Ecosystem W3C VC / DID Ecosystem Emerging Gap
Authentication Strong Limited Low
API Authorization Strong Weak Medium
Delegated Access Strong Partial Medium
Portable Credentials Partial Strong High
Machine/Agent Identity Emerging Emerging High
Enterprise Federation Strong Weak Medium
Reputation / Provenance Weak Partial High
Metadata Discovery OAuth metadata DID resolution Converging
Selective Disclosure Emerging (SD-JWT) Strong Narrowing
Operational Deployment Mature Fragmented High

The key strategic insight is that AI agents are collapsing the historical
separation between identity, authorization, credentials, and trust
infrastructure. This is forcing convergence between ecosystems that
previously evolved independently and sometimes antagonistically. The most
successful future standards are therefore unlikely to be exclusively
“OAuth-native” or exclusively “DID-native.” Instead, they will probably
combine OAuth-grade operational interoperability with verifiable credential
portability and cryptographic trust assertions. Projects such as auth.md
are early indicators of this convergence pattern, even if they currently
position themselves primarily within the IETF/OAuth lineage rather than the
W3C identity ecosystem.


On Mon, May 25, 2026 at 10:28 PM Aaron Parecki <aaron@parecki.com> wrote:

> This is based on this OAuth draft:
>
>
> https://datatracker.ietf.org/doc/draft-ietf-oauth-identity-assertion-authz-grant/
>
>
>
>
> On Mon, May 25, 2026 at 7:22 AM Paola Di Maio <paola.dimaio@gmail.com>
> wrote:
>
>> Greetings everyone
>>
>> I am simply keeping an eye out on thiings and have come across this
>> https://github.com/workos/auth.md
>>
>> I wonder, how does this relate to everything else we are
>> discussing/doing/observing in this space?
>>
>> My understanding is that  this is an emerging protocol proposal that may
>> be relevant to ongoing discussions around federated identity, delegated
>> authorization, verifiable credentials, and autonomous software agents  but
>> not a W3C thing
>> Please feel free to share with related CGs
>>
>> brief analysis:
>>
>>  “auth.md”, an OAuth-based agent registration and authorization approach
>> intended for autonomous AI agents operating without traditional
>> browser-mediated consent flows.
>>
>> What makes this notable is not necessarily the specific proposal itself,
>> but the architectural gap it exposes:
>>
>> Current OAuth/OIDC assumptions are heavily browser- and human-centric:
>>
>>    -
>>
>>    redirect-based consent
>>    -
>>
>>    interactive authorization
>>    -
>>
>>    session-oriented mediation
>>    -
>>
>>    user-present trust boundaries
>>
>> Autonomous agents introduce different requirements:
>>
>>    -
>>
>>    long-lived delegated authority
>>    -
>>
>>    non-interactive authorization
>>    -
>>
>>    machine-verifiable delegation chains
>>    -
>>
>>    portable trust assertions
>>    -
>>
>>    agent-scoped identity distinct from user identity
>>    -
>>
>>    policy-constrained autonomous execution
>>
>> The proposal appears to combine:
>>
>>    -
>>
>>    OAuth Protected Resource Metadata
>>    -
>>
>>    signed identity assertions
>>    -
>>
>>    machine-readable discovery metadata
>>    -
>>
>>    delegated authorization semantics for agents
>>
>> This seems highly adjacent to:
>>
>>    -
>>
>>    FedID discussions around federated assertions and browser mediation
>>    -
>>
>>    VC/DID work on portable cryptographic identity
>>    -
>>
>>    emerging “agent identity” and “agent authorization” efforts across
>>    OpenID, DIF, and IETF communities
>>
>> One particularly interesting aspect is the use of discoverable metadata
>> (“auth.md”) as a capability advertisement layer for agent onboarding and
>> authorization.
>>
>> I suspect we are beginning to see a broader standards gap emerge between:
>>
>>    -
>>
>>    “users using software”
>>    and
>>    -
>>
>>    “software acting autonomously under delegated authority”
>>
>> Questions I think may be worth discussing:
>>
>>    -
>>
>>    Are OAuth/OIDC extensions sufficient for agent-native delegation?
>>    -
>>
>>    Should agent identity be modeled independently from user identity?
>>    -
>>
>>    What role should VC/DID infrastructure play in portable agent trust?
>>    -
>>
>>    How should revocation and policy constraints operate for long-lived
>>    autonomous agents?
>>    -
>>
>>    Do we need standardized discovery metadata for agent authorization
>>    capabilities?
>>
>> Relevant references:
>>
>>    -
>>
>>    https://workos.com/auth-md
>>    -
>>
>>    https://workos.com/blog/agent-registration-with-auth-md
>>    -
>>
>>    https://www.w3.org/groups/wg/fedid/
>>    -
>>
>>    https://www.w3.org/community/credentials/
>>
>> Curious whether others see this as:
>>
>>    -
>>
>>    an implementation detail,
>>    -
>>
>>    an OAuth extension opportunity,
>>    or
>>    -
>>
>>    the beginning of a broader agent identity/authz standardization
>>    problem.
>>
>>
>>

Received on Wednesday, 27 May 2026 12:11:20 UTC