Independent Submission L. Muscariello Internet-Draft R. Polic Intended status: Informational Cisco Expires: 6 October 2025 4 April 2025 Agent Directory Service draft-agntcy-ads-latest Abstract The Agent Directory Service (ADS) is a distributed directory service designed to store metadata for AI agent applications. This metadata, stored as directory records, enables the discovery of agent applications with specific skills for solving various problems. The implementation features distributed directories that interconnect through a content-routing protocol. About This Document This note is to be removed before publishing as an RFC. The latest revision of this draft can be found at https://spec.dir.agncty.org. Status information for this document may be found at https://datatracker.ietf.org/doc/draft-agntcy-ads/. Source for this draft and an issue tracker can be found at https://github.com/agntcy/dir. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at https://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on 6 October 2025. Copyright Notice Copyright (c) 2025 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/ license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Table of Contents 1. Conventions and Definitions 2. Introduction 3. Naming 4. Content Routing 4.1. Capability Announcement 4.2. Discovery Process 4.3. Distributed Object Storage 4.3.1. Simplified Content Retrieval 4.3.2. OCI Integration 5. IANA Considerations 6. References 6.1. Normative References 6.2. Informative References Authors' Addresses 1. Conventions and Definitions The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here. 2. Introduction The Agent Directory Service (ADS) provides a mechanism for discovering AI agent applications based on their capabilities. It uses a distributed directory architecture to map agent skills to directory record identifiers and maintains a list of directory servers hosting those records. Directory records are identified by globally unique names that are routable within a DHT (Distributed Hash Table) to locate peer directory servers. The skill taxonomy is also routable in the DHT to map skillsets to records that announce those skills. Each directory record includes skills from a defined taxonomy, as specified in the [Taxonomy of AI Agent Skills] from [OASF]. While all record data is modeled using [OASF], only skills are leveraged for content routing in the distributed network of directory servers. This document describes the core concepts and architecture of the Agent Directory Service. 3. Naming In distributed systems, a reliable and collision-resistant naming scheme is crucial. The agent directory uses cryptographic hashes [RFC7838] to generate globally unique identifiers for data records. ADS leverages OCI as object storage, and therefore identifiers are made available as described in [OCI digest]. 4. Content Routing ADS implements capability-based record discovery through a hierarchical skill taxonomy. This architecture enables: 4.1. Capability Announcement Multi-agent systems publish their capabilities by encoding them as skill taxonomies. Each record contains metadata describing the agent's functional abilities. Skills are structured in a hierarchical format for efficient matching. 4.2. Discovery Process The system performs a two-phase discovery operation: 1. Matches queried capabilities against the skill taxonomy to determine records by their identifier 2. Identifies the server nodes storing relevant records. ## Distributed Resolution Local nodes execute targeted retrievals based on: 1. Skill matching results: Evaluates capability requirements. 2. Server location information: Determines optimal data sources. ADS uses libp2p [Kad-DHT] [DHT] for server and content discovery. +----------------+ | DHT Node | | Content Index | +----------------+ ^ | +----------------+-----------------+ | | | +-------v------+ +------v-------+ +-----v--------+ | Server Node A | | Server Node B| | Server Node C| | Content X | | Content Y | | Content Z | +-------+------+ +------+-------+ +------+-------+ | | | | Content Exchange | +---------------+------------------+ | Content Replication Flow: 1. Servers register content with DHT 2. DHT maintains content-to-server mappings 3. Servers query DHT to locate content 4. DHT returns list of servers hosting content 5. Servers download content from peers 4.3. Distributed Object Storage ADS differs from block storage systems like [IPFS] in its approach to distributed object storage. 4.3.1. Simplified Content Retrieval 1. ADS directly stores complete records rather than splitting them into blocks. 2. No special optimizations are needed for retrieving content from multiple sources. 3. Records are retrieved as complete units using standard OCI protocols. 4.3.2. OCI Integration ADS leverages the OCI distribution specification for content storage and retrieval: 1. Records are stored and transferred using OCI artifacts. 2. Any OCI distribution-compliant server can participate in the network. 3. Servers retrieve records directly from each other using standard OCI protocols. While ADS uses zot as its reference OCI server implementation, the system works with any server that implements the OCI distribution specification. 5. IANA Considerations This document has no IANA actions. 6. References 6.1. Normative References [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997, . [RFC6920] Farrell, S., Kutscher, D., Dannewitz, C., Ohlman, B., Keranen, A., and P. Hallam-Baker, "Naming Things with Hashes", RFC 6920, DOI 10.17487/RFC6920, April 2013, . [RFC7838] Nottingham, M., McManus, P., and J. Reschke, "HTTP Alternative Services", RFC 7838, DOI 10.17487/RFC7838, April 2016, . [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, May 2017, . 6.2. Informative References [DHT] "*** BROKEN REFERENCE ***". Authors' Addresses Luca Muscariello Cisco Email: lumuscar@cisco.com Ramiz Polic Cisco Email: rpolic@cisco.com