# Discovery of Agents With Names (DAWN)

Deployment of some AI systems requires interactions between a client and other AI resources (including but not limited to an AI agent). For some of these interactions to work, the client may need to discover some specific AI resources and learn how to contact them.

The focus of DAWN is on how a client can discover AI resource specific properties to facilitate subsequent interactions such as selection, invocation, or communication. Concretely, the WG will specify the minimum discovery information (e.g., type, reachability information, communication protocol, services). 

Note that discovery may be an iterative process, and may utilize either direct transfer or indirection.

To support these goals, the working group will specify an interoperable and generic discovery mechanism that builds on existing protocols and tools, benefits from established trust models, supports proven delegation and federation architectures, and allows organizations to have independently published discovery information.

## Scope 

Discovery in the DAWN context is limited to a client requesting, "Find me an agent resource to interact with" within a local network, within an organization, or between cooperating organizations with trust relationships over the public Internet.

Where possible, any solutions created by this WG will be built in a modular way using existing IETF protocols that provide support for any needed communication, authentication, and privacy. The WG will consider mDNS and DNS as potential mechanisms upon which to build a discovery protocol, but may examine other solutions as well.

The WG work items will provide integrity of all the discovery information exchanged between the DAWN parties except on the local network.

Although initially focused on discovery of AI agents and their capabilities, the WG should strive to produce results that are general and reusable within other discovery contexts if possible.

## Out of Scope

Specifically out of scope for the DAWN WG during its initial phase are:

- Exchange of capability information beyond the minimum discovery information as specified in the requirements section of the Architecture document.
- All communication beyond the initial discovery process.
- AI resource indexing across the wider Internet.
- Identity management of AI agents, tools and skills.
- Trust management and trust evaluation methods beyond secure content exchanged between DAWN parties.

Future DAWN or other WG charters may take on these tasks.

## Deliverables

The DAWN working group will work on the following deliverables:

- Terminology: The definition of common DAWN terminology for use in the other DAWN documents. (May be published as an Informational RFC or included in the Architecture document.)

- Use Cases and Applicability: A document describing broad DAWN use case categories. (May be published as an Informational RFC)

- Discovery Architecture: One or more documents describing the problem space, including deployment models (within a local network, within an organization, or between organisation), requirements (including the definition of the minimum set of required information about the AI resource and communication), and the resulting DAWN architecture. (To be published as Informational RFC(s), but maybe held until the Protocol work is considered stable.)

- Protocol: One or more specifications describing the implementations of the Discovery Architecture(s) (To be published as Standards Track document(s)).

## Coordination with Other Working Groups and Organizations

The DAWN working group will seek to coordinate with WIMSE and OAUTH WG as well as other WGs relating to DNS, discovery, identities, or AI. It will also coordinate with external standards bodies as necessary.