MapGraphSeamsCalendarIntroductionDisclosureMethodFor agents and tools

Updated weekly, new versions on Mondays · Corrections: open an issue

IETF DAWN, Discovery of Agents With Names (proposed WG)

IETFchartercore entry

Verified 2026-10-14. Status as of 2026-10-14. All entries.

Seams: S1

Status

Charter 00-10 has been in external review since 2026-10-09, with comments to the Internet Engineering Steering Group (IESG) due by 2026-10-19 and an IESG telechat listed for 2026-10-22; the Working Group (WG) is "Proposed", and the WG Review announcement of 2026-10-09 says "The IESG has not made any determination yet". The Approve ballot on 00-10, read on the datatracker on 2026-10-14, carries the summary "Has enough positions to pass." with one Yes and one No Objection recorded; the charter has not been recorded as approved, and no WG Action announcement for DAWN is found among the ietf-announce messages up to 2026-10-12 (read 2026-10-13). The announcement names the chairs of the proposed WG: Adrian Farrel and Wes Hardaker. Earlier history: charter 00-09 was on the IESG ballot: on 2026-10-06 one IESG member changed from Block (on 00-08) to No Objection and another voted Yes; on 2026-10-07 one member voted Yes with comments and three recorded No Objection; on 2026-10-08 two more recorded No Objection; the IESG telechat had been set for 2026-10-08. On 2026-10-09 the charter was revised to 00-10 and sent to external review: the datatracker history of that day shows the state changed from "Start Chartering/Rechartering (Internal Steering Group/IAB Review)" to "External Review (Message to Community, Selected by Secretariat)", the "Ready for external review" ballot closed, an "Approve" ballot created and the telechat date changed from 2026-10-08 to 2026-10-22; one IESG member recorded Yes on the new ballot on 2026-10-12. The datatracker pages read on 2026-10-13 show the WG as "Proposed" and charter-ietf-dawn-00-10 in that state, with "On agenda of 2026-10-22 IESG telechat". The Secretariat's announcement "WG Review: Discovery of Agents With Names (dawn)", sent to ietf-announce on 2026-10-09, says "The IESG has not made any determination yet" and asks for comments to the IESG mailing list by 2026-10-19. No WG Action announcement for DAWN is among the ietf-announce messages up to 2026-10-12, the latest of the list on 2026-10-13; a decision on DAWN is not found.

Relevance

DAWN would standardise how a client discovers "public properties of AI resources", so it is the IETF venue where a seller's agent surface would be found. The charter gives three deployment contexts with one example each (charter 00-10, read on the datatracker on 2026-10-14): for the local network "within the local network, e.g., discover all agents in this shop;", and for the public Internet "between cooperating parties with trust relationships over the public Internet, e.g., discover all agents of example.org, for all its customers/members.". The WG "will also specify the minimum public discovery information (e.g., type, reachability information, communication protocol, services)."; commerce as a scope item: not found in charter 00-10, read 2026-10-13 (commerce, merchant, transaction, payment and purchase do not occur; "shop" occurs once, in the local network example "within the local network, e.g., discover all agents in this shop;"), nor in charter 00-08, 2026-10-03.

On 2026-10-05 the INT Area Director wrote, on a request for a non-WG mailing list for DNS records under underscore labels that carry keys, key commitments or references to them, "While a mailing list will be created, I would like to push back on its name as it misses link to DNS"; the requester proposed the name dns-underscore-keys.

An individual survey, draft-jimenez-dawn-discovery-landscape-00 (2026-07-03), compares discovery mechanisms from the IETF, AAIF, 3GPP and open-source projects and lists eight gaps, among them "Well-Known Namespace Pressure"; a text search of it finds no occurrence of AI Catalog, Agentic Resource Discovery (ARD), UCP or commerce (2026-10-06). draft-zhang-dawn-agent-discovery-framework-01 (2026-09-06) proposes a two-layer federated architecture.

A second individual draft, draft-moussa-dawn-gap-analysis-01 (Moussa, Akhavain, 2026-06-12), is a gap analysis and applicability statement that maps existing discovery mechanisms to the DAWN problem statement and requirements; a text search of it finds no occurrence of AI Catalog, ARD, UCP or commerce (2026-10-06). draft-kay-dawn-use-cases-01 (2026-09-10) names Agentic Resource Discovery as a possible interface to a provider's catalogue in a broadcast use case and adds: "ARD is an illustrative option; protocol choice remains open." The IETF Chair announced on 2026-10-02 that an INT Area Director resigned from the IESG as of 2026-10-01 and that the other co-AD for the INT area is the primary point of contact for the area; the datatracker lists that Area Director as responsible Area Director (AD) for DAWN (2026-10-03).

On 2026-10-07 an IESG member entered a Yes with comments on 00-09 that asks the charter to say that the protocol specifications normatively define the minimum discovery information, that the WG is "expected to converge on one mechanism per deployment context", and that "discovering where a catalog is lives in DAWN, and what a catalog contains doesn't"; the member also asks whether DAWN defines how agents are named, and that DNSSD, DNSOP and AgentProto be named in the coordination text.

On the list thread "DAWN Coordination: Discovery Information and Trust Artifact Pointers" Girish Konda wrote on 2026-10-07 that a reference that does not resolve, one blocked by the discovering entity's own retrieval policy, and one that resolves with the record absent are three causes: "All three end as no discovery result, and they have different owners and different retries." On 2026-10-12 a participant who states that they are a co-author of three individual drafts in this space wrote to the IESG on the DAWN list in reply to the WG Review thread "I support chartering DAWN as written in charter-ietf-dawn-00-10", and gave as a reason for keeping identity management of agents out of the first phase "Discovery has to work before anyone argues about who the discovered thing is".

Committed next

Boundary

Out of scope in the initial phase, charter 00-10 (read 2026-10-13): catalogs of capabilities, all communication beyond the initial discovery process, "AI resource indexing and searching across the wider Internet", "Identity management of AI agents, tools, and skills." and trust management beyond secure content exchange between DAWN parties. In scope as a discoverable AI resource: "a service such as a catalog that will further the discovery process".

Author participation

The author of this map wrote in the DAWN list thread "DAWN Coordination: Discovery Information and Trust Artifact Pointers", the latest messages on 2026-09-29, on 2026-10-08 (replying to Girish Konda) and on 2026-10-10 (accepting Konda's wording on absence and incomplete results); the reply of Konda adds a third case to the author's second requirement; see Disclosure.

Sources

Overlaps

W3C Agentic Web Working Group (draft charter), Agentic Resource Discovery (ARD), A2A protocol, agent discovery

Neighbours in the map

Neighbours of IETF DAWN, Discovery of Agents With Names (proposed WG) in the map Each circle is an entry of the map and links to its page; the colour is the kind of body behind it, a filled circle is a core entry and a dashed circle a context entry. The entry in the middle, with the entries it overlaps and the entries named by the same seams. Overlaps (6)Share a seam (11)W3C Agentic Web Working Group (draft charter) (W3C, core)W3C Agentic Web Working… IETF agentproto (Agent Communication Protocols WG) (IETF, core)IETF agentproto Individual IETF drafts on the seller side (IETF, core)Individual IETF drafts… IEEE P3931 Agent Description, Discovery, and Registry (ADDR) Interoperability (Other standards bodies, context)IEEE P3931 Agent Descri… A2A protocol, agent discovery (Linux Foundation, core)A2A protocol Agentic Resource Discovery (ARD) (Independent, core)Agentic Resource Discov… W3C AI Agent Protocol Community Group (W3C, core)W3C AI Agent Protocol C… S1 DNS for AI Discovery (DNS-AID, draft-mozleywilliams-dnsop-dnsaid) (IETF, context)DNS for AI Discovery S1 AI Catalog (Agent Card Working Group) (Linux Foundation, core)AI Catalog S1 Agent2Agent (A2A) protocol, core specification (Linux Foundation, context)Agent2Agent S1 MCP Server Cards (ext-server-card, SEP-2127) (Linux Foundation, core)MCP Server Cards S1 Model Context Protocol (MCP), core specification (Linux Foundation, context)Model Context Protocol S1 schema.org vocabulary for products, offers, sellers and lodging (current release) (Independent, context)schema.org vocabulary f… S1 Personal Agent Consent & Trust Protocol (PACT), published by Decagon with Instinct (Industry, core)Personal Agent Consent… S1 Personal Agent Protocol (PAP, "Poppy"), published by Sierra with Meta (Industry, core)Personal Agent Protocol S1 Shopify, MCP surfaces for buyer agents, developers and merchants (Industry, context)Shopify S1 Universal Commerce Protocol (UCP) (Industry, core)Universal Commerce Prot… S1 IETF DAWN

Solid lines: overlaps declared by one of the two entries. Dashed lines, coloured by seam: entries named by the same seam (S1 to S4) in the seams note. Every node is a link.

Evidence and coverage

Read
On 2026-10-14: the datatracker charter page and the ballot page of charter-ietf-dawn-00-10 (summary, positions) and the WG Review announcement (chairs).
Not read
Ballot comments and positions beyond the summary and the count of positions, the ietf-announce messages after 2026-10-12, the DAWN list messages of 2026-10-10 to 2026-10-12 other than the one cited.
Not found
Not found in the 200 messages, 2026-10-03: AI Catalog.

Open proposals

ProposalAuthorDateStateLink
First work item: an organisation's agent index, catalogue or registry. Against charter 00-10 (read 2026-10-13): the charter lists "a service such as a catalog that will further the discovery process" as a discoverable AI resource and excludes catalogs of capabilities and "AI resource indexing and searching across the wider Internet"; the proposal is not among its work items.Jim Mozley2026-08-05openhttps://mailarchive.ietf.org/arch/msg/dawn/fsJ8oAdR8rpNPhCj6LP9sspbxBA