<?xml version="1.0" encoding="US-ASCII"?>
<!-- <?xml version="1.0" encoding="UTF-8"?> -->
<!-- edited with XMLSPY v5 rel. 3 U (http://www.xmlspy.com)
     by Daniel M Kohn (private)
-->
<!DOCTYPE rfc SYSTEM "rfc2629.dtd">


<?rfc toc="yes"?>
<?rfc rfcedstyle="yes"?>
<?rfc subcompact="no"?>
<?rfc symrefs="yes"?>

<rfc ipr="trust200902" category="info" docName="draft-jeong-nmrg-agentic-ai-framework-01">

<front>
    <title abbrev="Agentic AI Framework">
    An Agentic AI Framework for the Management of Networks, Security, and AI Services
    </title>

    <author role="editor" initials="J." surname="Jeong" fullname="Jaehoon Paul Jeong">
        <organization abbrev="Sungkyunkwan University">
        Department of Computer Science &amp; Engineering
        </organization>
        <address>
            <postal>                
                <extaddr>Sungkyunkwan University</extaddr>
                <street>2066 Seobu-Ro, Jangan-Gu</street>
                <city>Suwon</city> <region>Gyeonggi-Do</region>
                <code>16419</code>
                <country>Republic of Korea</country>
            </postal>
            <phone>+82 31 299 4957</phone>
            <email>pauljeong@skku.edu</email>
            <uri>http://iotlab.skku.edu/people-jaehoon-jeong.php
         </uri>
        </address>
    </author>

    <author initials="M." surname="Gu" fullname="Mose Gu">
        <organization abbrev="Sungkyunkwan University">
        Department of Computer Science &amp; Engineering
        </organization>	
		<address>
		    <postal>
			    <extaddr>Sungkyunkwan University</extaddr>
  			    <street>2066 Seobu-Ro, Jangan-Gu</street>
				<city>Suwon</city>
				<region>Gyeonggi-Do</region>
				<code>16419</code>
				<country>Republic of Korea</country>
			</postal>
			<phone>+82 31 299 4106</phone>
		    <email>rna0415@skku.edu</email>
			<uri>http://iotlab.skku.edu/people-Moses-Gu.php</uri>
		</address>
    </author>

    <author initials="Y." surname="Ahn" fullname="Yoseop Ahn">
        <organization abbrev="Sungkyunkwan University">
        Department of Computer Science &amp; Engineering
        </organization>	
		<address>
		    <postal>
			    <extaddr>Sungkyunkwan University</extaddr>
  			    <street>2066 Seobu-Ro, Jangan-Gu</street>
				<city>Suwon</city>
				<region>Gyeonggi-Do</region>
				<code>16419</code>
				<country>Republic of Korea</country>
			</postal>
			<phone>+82 31 299 4106</phone>
		    <email>ahnjs124@skku.edu</email>
			<uri>http://iotlab.skku.edu/people-Ahn-Yoseop.php</uri>
		</address>
    </author>

    <author initials="J." surname="Park" fullname="Jung-Soo Park">
      <organization abbrev="ETRI">
      Electronics and Telecommunications Research Institute
      </organization>
      <address>
        <postal>
          <street>218 Gajeong-Ro, Yuseong-Gu</street>
          <city>Daejeon</city>
          <code>34129</code>
          <country>Republic of Korea</country>
        </postal>
        <phone>+82 42 860 6514</phone>
        <email>pjs@etri.re.kr</email>
      </address>
    </author>

    <author initials="B." surname="An" fullname="Byoungman Robert An">
        <organization abbrev="KETI">
        Intelligent Information R&amp;D Division
        </organization>
        <address>
            <postal>
                <extaddr>Korea Electronics Technology Institute</extaddr>
                <street>Global R and D Center 6th floor</street>
                <street>#22, Daewangpangyo-ro 712beon-gil</street>
                <city>Seongnam</city> <region>Gyeonggi-Do</region>
                <code>13488</code>
                <country>Republic of Korea</country>
            </postal>
            <phone>+82 31 739 7463</phone>
            <email>bman@keti.re.kr</email>
            <uri>https://www.keti.re.kr/eng/main/main.php
         </uri>
        </address>
    </author>

    <date month="November" day="16" year="2026" />

    <area>Network Management</area>
    
    <workgroup>Internet Research Task Force</workgroup>

<!-- [rfced] Please insert any keywords (beyond those that appear in
the title) for use on http://www.rfc-editor.org/rfcsearch.html. -->

    <keyword>
    Agentic AI Framework, A2A, MCP, Networks, Security, AI, Services,
    Management, Agent, Agent Card, Intent
    </keyword>


    <abstract>
        <t>
        This document specifies an agentic AI framework for the management of
        networks, security, and AI services.  This framework is an
        Intent-Based System (IBS) to enforce a human-language intent at a
        target network with the translated policy for networks, security, or
        AI services.  It consists of a User System, a Cloud System, an Edge
        System, and a Device.  Target devices in this framework include
        Network Forwarding Elements (e.g., router and switch), Network
        Security Functions (e.g., firewall, web filter, and DDoS-attack
        mitigator), Internet-of-Things (IoT) Devices, Physical AI Robots,
        Software-Defined Vehicles (SDV), Unmanned Aerial Vehicles (UAV), and
        Urban Air Mobility (UAM).  This document describes the agentic AI 
        framework with its components and interfaces where the components
        include AI agents and supporting knowledge repositories (i.e., 
        knowledge base), and also inter-agent interfaces use the
        Agent-to-Agent (A2A) protocol.
        </t>
    </abstract>
</front>

<middle>

<section anchor="section:Introduction" title="Introduction">
    <t>
    Agentic AI opens a new door providing human users with human-friendly
    services through Large Language Models (LLMs) such as GPT-3 <xref target="GPT-3"/>
    and Flan-T5 <xref target="Flan-T5"/>.
    Also, network softwarization has been widely adopted in various domains
    such as data centers and mobile networks <xref target="TS-23.501" />.
    The network softwarization is realized with Network Functions
    Virtualization (NFV) <xref target="ETSI-NFV" /><xref
    target="ETSI-NFV-Release-2" /> and Software-Defined Networking (SDN) <xref
    target="RFC7149" />. Furthermore, Intent-Based Networking (IBN) <xref target="RFC9315" /><xref
    target="Survey-IBN-CST-2023" /><xref target="ClickINC" /> can support
    human-friendly configuration for target networks and security systems.
    </t>

    <t>
    This document specifies an agentic AI framework for the management of
    networks, security, and AI services.  This framework is an
    Intent-Based System (IBS) including the IBN that enforces a
    human-language intent at a target network. It aims at the IBS in data
    centers, corporate networks, campus networks, and home networks rather
    than public networks like the Internet. 
    </t>

    <t>
    The agentic AI framework in this document supports the translation of
    a human-language intent and the enforcement of the corresponding
    service(s) through multiple AI agents in target networks.  This intent
    can be translated into a specific policy for networks, security, or AI
    services.  The agentic AI framework in this document consists of a User
    System, a Cloud System, an Edge System, and a Device.  Devices under the
    control of this framework include Network Forwarding Elements (e.g.,
    router and switch), Network Security Functions (e.g., firewall, web
    filter, and DDoS-attack mitigator), Internet-of-Things (IoT) Devices,
    Physical AI Robots, Software-Defined Vehicles (SDV) 
    <xref target="AUTOSAR-SDV" /><xref target="Eclipse-SDV" /><xref
    target="COVESA" />, Unmanned Aerial Vehicles (UAV), and Urban Air
    Mobility (UAM).  Thus, this document describes the agentic AI framework
    with its components and interfaces where the components are AI agents and
    the interfaces use the Agent-to-Agent (A2A) protocol
    <xref target="A2A"/><xref target="A2A-Specification"/>.
    </t>
</section>

<section anchor="section:Terminology" title="Terminology">
    <t>
    This document uses the terminology described in <xref target="RFC9315" />
    and <xref target="I-D.irtf-nmrg-ibn-usecases"/>.  The new terms are
    defined below:
    </t>

    <t>
    <ul>
      <li>
        AI Agent: A software module that can translate a human-language
        intent into a structured policy in  a Domain-Specific Language (DSL).
        An AI agent can communicate with another AI agent for collaboration
        via the Agent-to-Agent (A2A) protocol.  It can use available
        computing, storage, and network resources via Model Context Protocol
        (MCP) <xref target="MCP"/><xref target="MCP-Specification"/>. The MCP
        accesses data and file systems, development tools, productivity tools,
        and physical AI tools for the requested service.
      </li>

      <li>
        Agent Catalog: A logical collection of Agent Cards
        <xref target="A2A-Specification"/>, or references to them, describing
        AI agents known to the framework.  It supports the lookup of
        candidate AI agents based on their advertised skills and interaction
        requirements.  In this document, the Agent Catalog constitutes the
        agent-metadata part of the Knowledge Base that is a database for
        intent translation.  Entries in the Agent Catalog may be populated
        and refreshed through discovery or registration mechanisms.  A
        storage representation format (e.g., 
        XML <xref target="W3C.REC-xml-20081126"/>,
        JSON <xref target="RFC8259"/>, or 
        YAML <xref target="YAML"/>), access
        information (e.g., IP address and port number), and an access
        mechanism may be included in the Agent Catalog.
      </li>

      <li>
        Intent: A set of operational goals (i.e., what a network should
        meet) and outcomes (i.e., what the network is supposed to deliver)
        defined in a declarative manner without specifying how they are
        achieved or should be implemented <xref target="RFC9315" />.
      </li>

      <li>
        Intent-Based System (IBS): A system that enforces an intent from a
        user (or administrator) into a target system (e.g., router, switch, 
        firewall, web filter, IoT device, robot, and SDV) through an intent
        life cycle <xref target="RFC9315"/>. 
      </li>

      <li>
        Intent Translator: A component that can translate an intent in a
        natural language (e.g., English) into a specific intent (i.e., a
        refined intent) or the corresponding high-level policy (i.e., a
        network policy, security policy, or AI service policy) using
        Natural Language Processing (NLP) and Large Language Models (LLMs)
        <xref target="Flan-T5"/><xref target="GPT-3"/>
        <xref target="USENIX-ATC-Lumi" /><xref target="BERT" />.
        That is, an Intent Translator converts a natural-language intent
        into a structured policy representation (i.e., a specific intent or a
        high-level policy) based on <xref target="I-D.gu-nmrg-intent-translator"/>.
        A specific intent denotes a refined, structured representation of a goal
        assigned to a supervised agent (e.g., Edge Agent or Worker Agent) or
        a high-level policy with concrete rules.  It also selects at least one
        supervised agent to enforce the specific intent or the high-level policy.
      </li> 

      <li>
        Knowledge Base (KB): A logical repository that provides
        domain-specific policy knowledge and agent metadata used
        by the framework. Policy knowledge includes target schemas,
        valid policy examples, and semantic constraints, drawing
        on the Policy Knowledge Base concept described in
        <xref target="I-D.gu-nmrg-intent-translator"/>.
        In this document, agent metadata is organized as an
        Agent Catalog. Policy knowledge and agent metadata are
        distinct logical categories and may reside in separate
        physical repositories. A graph-based representation is
        an implementation option.
      </li>

      <li>Orchestrator: A component that governs and monitors supervised
        agents (e.g., Edge Agent and Worker Agent) to achieve the fulfillment
        and assurance of the target intent <xref target="RFC9315" />.   
        Through the monitoring of the supervised agents by the Orchestrator,
        the activities and performance of those agents are monitored and
        analyzed.  If needed, the specific intent is augmented by the
        Orchestrator.  Also, in some cases, a new specific intent may be
        generated by the Orchestrator and be delivered to an appropriate
        target agent for reconfiguration.
      </li>

      <li>
        Physical AI Object (PAO): A physical object that can move with an AI agent,
        power source, and wireless communication capability (e.g., 5G
        Vehicle-to-Everything (V2X)). A PAO can be an intelligent
        Internet-of-Things (IoT) Device, Software-Defined Vehicle (SDV)
        <xref target="AUTOSAR-SDV"/><xref target="Eclipse-SDV" /><xref target="COVESA" />,
        Unmanned Aerial Vehicle (UAV), or Urban Air Mobility (UAM).
      </li>

      <li>
        Virtual AI Object (VAO): A virtual object that can handle network
        and security services with an AI agent and wired/wireless 
        communication capability (e.g., Gigabit Ethernet, Wi-Fi, and 5G V2X). A VAO
        can be either an SDN Controller <xref target="RFC7149" /> for network
        management or a Security Controller
        <xref target="I-D.ietf-i2nsf-applicability"/> for security management.
      </li>
    </ul>
    </t>

</section>

<section title="Agentic AI Framework for Service Management">
    <t>
    This section specifies an Agentic AI System for AI Service Management in
    target domains, as shown in <xref target="figure:Agentic-AI-System-for-Service-Management" />.
    In this figure, a human user (e.g., network administrator, building 
    manager, and home user) uses a User System (e.g., smartphone and
    computer) for managing networks, security, and AI services.  The User
    System interacts with a Cloud System or an Edge System for the human
    user's intents.  A Cloud System manages multiple Edge Systems for the
    sake of human users. An Edge System manages multiple Devices for
    networks, security, and AI services like in 
    <xref target="figure:Agentic-AI-System-for-Service-Management" />.
    The devices can be either Virtual AI Objects (VAOs) or Physical AI
    Objects (PAOs) having at least one Worker AI Agent for the interaction
    with an Edge Agent and other Worker AI Agents.
    </t>

    <figure anchor="figure:Agentic-AI-System-for-Service-Management" align="center"
      title="Agentic AI System for Network, Security, and AI Service Management">
        <artwork align="left"><![CDATA[
                                                User System
                                               +------------+
                                               | User Agent |
                                               +------------+
                                               ^            ^
                                               | A2A        | A2A
                                               |            v
                        Edge System            v         Cloud System  
        ******************************************        **********
       *                                          *     *            *
      *                                            *   * +----------+ *
      *             +-----------------+            *   * |  Cloud   | *
      *             |   Edge Agent    |            *<->* |  Agent   | *
      *             +--------^--------+            *A2A* +----------+ *
      *                      |                     *   *              *
       *                     v                    *     *            *
        ******************************************        **********
        ^                    ^                   ^
        | A2A                | A2A               | A2A
        v                    v                   v
+----------------+    +----------------+    +----------------+
|   +------+     |    |   +------+     |    |   +------+     |
|   |Worker|     |    |   |Worker|     |    |   |Worker|     | 
|   |Agent1|     |<-->|   |Agent2|     |<-->|   |Agent3|     |
|   +---^--+     |A2A |   +---^--+     |A2A |   +---^--+     | 
|       | MCP    |    |       | MCP    |    |       | MCP    |
|    +--v----+   |    |    +--v----+   |    |    +--v----+   |
|    |Modules|   |    |    |Modules|   |    |    |Modules|   |
|    +-------+   |    |    +-------+   |    |    +-------+   |
+----------------+    +----------------+    +----------------+     
     Device1               Device2               Device3 

<----> Wired/Wireless Link
]]></artwork>
      </figure>

    <t>
    <xref target="figure:Agentic-AI-Framework-for-Service-Management" />
    shows an Agentic AI Framework for the management of networks, security,
    and AI services.
    A User Agent in the User System generates an intent for the human user's
    service demand and sends it to either a Cloud System or an Edge System.
    Either a Cloud Agent in the Cloud System or an Edge Agent in the Edge
    System receives the intent from the User Agent.
    </t>

    <t>
    In the case where the Cloud Agent receives an intent from the User Agent,
    the Cloud Agent translates the intent into the corresponding high-level 
    policy through its Intent Translator. 
    For the authentication and authorization of a User Agent, the User Agent
    in the User System  needs to contact the Cloud Agent in the Cloud System.
    This security process is necessary for the management of networks,
    security, and AI services at the target domains where the User Agent has
    access permission.  This is different from the assumption that an AI
    agent can discover other AI agents to help its task like travel agents
    to reserve airplane tickets, hotel rooms, and local transportation
    tickets.      
    </t>

    <figure anchor="figure:Agentic-AI-Framework-for-Service-Management" align="center"
      title="Agentic AI Framework for Network, Security, and AI Service Management">
        <artwork><![CDATA[            
     <User System>
+----------------------+
|      User Agent      |
|(Smartphone, Computer)|
+----------------------+
 ^   |          ^
 |   | Intent   | Report
 |   v          |            <Cloud System>
 |  +----------------------------------------------------------------+
 |  | +-------------------+ Agent Information +--------------------+ |
 |  | |    Cloud Agent    |<----------------->|   Knowledge Base   | |
 |  | |(Intent Translator,|                   |(Edge Agent Catalog)| |
 |  | |   Orchestrator)   |                   |                    | |
 |  | +-------------------+                   +--------------------+ |
 |  +----------------------------------------------------------------+
 |         ^                                            ^
 | Intent  | Intent (Report, Monitoring Data)           | Agent Card
 |(Report) |                                            |
 v         v               <Edge System>                v
 +------------------------------------------------------------------+
 | +-------------------+                   +----------------------+ |
 | |    Edge Agent     | Agent Information |    Knowledge Base    | |
 | |(Intent Translator,|<----------------->|(Worker Agent Catalog)| |
 | |   Orchestrator)   |                   |                      | |
 | +-------------------+                   +----------------------+ |
 +------------------------------------------------------------------+
           ^                                            ^ 
           | Intent (Report, Monitoring Data)           | Agent Card
           |                                            |
           v          <Devices with Worker Agent>       v
+---------------------------------------------------------------------+
|      <Device1>               <Device2>               <Device3>      |
| +------------------+  +----------------------+  +-----------------+ |
| |  Worker Agent1:  |  |    Worker Agent2:    |  |  Worker Agent3: | |
| |  SDN Controller  |  | Security Controller  |  |   Physical AI   | |
| | (Router, Switch) |  |(Firewall, Web Filter)|  |(Robot, SDV, IoT)| |
| +------------------+  +----------------------+  +-----------------+ |
+---------------------------------------------------------------------+     
            ]]></artwork>
    </figure>
    
    <t>
    Each Cloud Agent or Edge Agent accesses its Knowledge Base through a
    repository access interface.  Its Intent Translator retrieves
    domain-specific schemas and semantic constraints from the
    policy-knowledge part of the Knowledge Base to support policy
    generation and validation. 
    There exists an Agent Catalog that is a collection of Agent Cards
    <xref target="A2A-Specification"/> used for searching for candidate
    agents to perform the intent.  There are two kinds of Agent Catalogs,
    namely an Edge Agent Catalog and a Worker Agent Catalog.  The Edge Agent
    Catalog contains the Agent Cards of Edge Agents inside Edge Systems.
    On the other hand, the Worder Agent Catalog contains the Agent Cards of
    Workder Agents inside Devices. 
    The Knowledge Base supplies information for intent translation and
    enforcement such as intent translation, policy validation, and agent
    selection.
    </t>

    <t>
    Note that the agent discovery for an Agent Catalog can be performed in
    the following two approaches.  The first approach is a reactive way using
    an agent discovery protocol (e.g., DNS and mDNS)
    <xref target="I-D.seethiraju-dawn-dan"/><xref target="I-D.zhang-dawn-agent-discovery-framework"/>.
    The second approach is a proactive way (e.g., a capability registration
    protocol) <xref target="I-D.ietf-i2nsf-capability-data-model"/><xref target="I-D.ietf-i2nsf-registration-interface-dm"/><xref target="I-D.ietf-i2nsf-applicability"/>.
    The Agentic AI Framework in this document can accommodate these two kinds
    of agent discovery approaches.
    </t>

    <t>
    A policy-knowledge role is used to translate an intent into 
    a specific intent (i.e., a refined intent) with the Policy Knowledge Base
    in <xref target="I-D.gu-nmrg-intent-translator"/>.
    Policy schemas and semantic constraints are supplied for each supported
    domain.  For policy generation, a schema (i.e., a data model) with
    constraints for the service management needs to be developed.
    Note that a policy based on such a scheme can be expressed in  
    XML <xref target="W3C.REC-xml-20081126"/>,
    JSON <xref target="RFC8259"/>, or 
    YAML <xref target="YAML"/>.
    </t>

    <t>
    Once the User Agent is authenticated and authorized through the Cloud
    Agent, the User Agent can directly interact with an Edge Agent in an
    Edge System that can handle the User Agent's intent with its Work Agents,
    as shown in <xref target="figure:Agentic-AI-Framework-for-Service-Management" />.
    </t>

    <t>
    When the Edge Agent receives a specific intent (i.e., a refined intent)
    from the Cloud Agent, it refines that intent into another specific intent
    or a high-level policy for Worker Agents.  When it receives a
    natural-language intent directly from the User Agent, its Intent
    Translator first generates a structured high-level policy, which is then
    refined in the same way.  Each specific intent represents a
    worker-specific goal or policy in an XML, JSON, or YAML document
    according to its target model.  Within this framework, the Intent
    Translator queries the Agent Catalog to obtain candidate Agent Cards and
    evaluates their suitability for the requested task.  Policy refinement and
    Worker Agent selection are extensions introduced by this document.
    Refinement preserves the goals and constraints of the original intent.
    </t>

    <t>
    The Edge Agent sends the specific intent or the high-level policy to the
    selected Worker Agent(s) to enforce the requested service corresponding
    to the intent from the User Agent.
    </t>

    <t>
    When a Worker Agent receives a specific intent or a high-level policy, it
    performs the requested service using its internal resources and external
    resources via MCP <xref target="MCP-Specification"/> with predefined APIs.
    </t>

    <t>
    Operation results for the given intent are delivered as a report
    (e.g., success or failure) or monitoring data (e.g., alarms, events, 
    logs, or counters)
    <xref target="I-D.ietf-i2nsf-nsf-monitoring-data-model" />
    from a supervised source component (e.g., Worker Agent, Edge Agent, and
    Cloud Agent) to a supervising destination component (e.g., Edge Agent,
    Cloud Agent, and User Agent) in the Agentic AI Framework in
    <xref target="figure:Agentic-AI-Framework-for-Service-Management" />.
    When a Cloud Agent receives a report or monitoring data from
    an Edge Agent, the Orchestrator in the Cloud Agent checks the report or
    monitoring data to determine an appropriate action, depending on the
    message in the report or the monitoring data.
    Also, when an Edge Agent receives a report or monitoring data from
    a Worker Agent, the Orchestrator in the Edge Agent handles the report or
    monitoring data in the same way as the Cloud Agent.
    Therefore, the operation results (i.e., reports and monitoring data) can
    be used as feedback for fulfilling and assuring the intent through the
    closed-loop control in the Agentic AI Framework
    <xref target="I-D.irtf-nmrg-ibn-usecases" />.
    </t>
 </section>

<section title="Interfaces in Agentic AI Framework">
    <t>
    Interfaces in the Agentic AI Framework use A2A <xref target="A2A-Specification"/>
    and MCP <xref target="MCP-Specification"/>.
    The interfaces among the User Agent, the Cloud Agent, the Edge Agent, and
    the Worker Agents use A2A, as shown in
    <xref target="figure:Agentic-AI-System-for-Service-Management" />.
    The message formats used by A2A conform to the specification of A2A
    <xref target="A2A-Specification"/>.
    The interfaces among agents and modules in each system (e.g., User Agent,
    Cloud Agent, Edge Agent, and Worker Agent) use MCP, as shown in
    <xref target="figure:Agentic-AI-System-for-Service-Management" />.
    The message formats used by MCP conform to the specification of MCP
    <xref target="MCP-Specification"/>.
    </t> 
</section>

<section anchor="section:IANA-Considerations" title="IANA Considerations">
  <t>
    This document does not require any IANA actions.
  </t>
</section>

<section anchor="section:Security-Considerations" title="Security Considerations">
  <t>
  Authentication and authorization mechanisms should be used to check the
  validity of User Agents.  For trustworthy cooperation, mutual
  authentication should be performed beforehand among the Cloud Agent, the
  Edge Agents, and Worker Agents.
  </t>

  <t>
  An intent from a User Agent should be evaluated by either the Cloud Agent
  or the Edge Agent to determine whether it is harmless to the Agentic AI
  Framework or not.  If the intent is harmful, either the Cloud Agent or the
  Edge Agent should log this intent and report the harmfulness of the intent
  to the User Agent which issued the intent.
  </t>

  <t>
  For reliable and agile AI services, the Agentic AI Framework should perform
  governance by monitoring and auditing the activities of AI agents including
  the User Agent, Cloud Agent, Edge Agents, and Worker Agents.  Such an
  auditing system may operate as an independent entity in the Agentic AI
  Framework so that it can audit the AI agents without any bias.
  </t>
</section>

</middle>

<back>

<!-- START: Normative References -->
<references title="Normative References">

    <?rfc include="reference.RFC.6020"?>
    <?rfc include="reference.RFC.6241"?>
    <?rfc include="reference.RFC.7149"?>
    <?rfc include="reference.RFC.7950"?>
    <?rfc include="reference.RFC.8040"?>
    <?rfc include="reference.RFC.8259"?>    
    <?rfc include="reference.RFC.8329"?>
    <?rfc include="reference.RFC.9315"?>
    <?rfc include="reference.RFC.9365"?>
    
</references>
<!-- END: Normative References -->

<!-- START: Informative References -->
<references title="Informative References">

    <?rfc include='reference.I-D.irtf-nmrg-ibn-usecases'?>
    <?rfc include='reference.I-D.gu-nmrg-intent-translator'?>
    <?rfc include='reference.I-D.seethiraju-dawn-dan'?>
    <?rfc include='reference.I-D.zhang-dawn-agent-discovery-framework'?>
    <?rfc include='reference.I-D.ietf-i2nsf-capability-data-model'?>
    <?rfc include='reference.I-D.ietf-i2nsf-registration-interface-dm'?>
    <?rfc include='reference.I-D.ietf-i2nsf-nsf-monitoring-data-model'?>
    <?rfc include='reference.I-D.ietf-i2nsf-applicability'?>

    <reference anchor="W3C.REC-xml-20081126" target="https://www.w3.org/TR/2008/REC-xml-20081126/">
      <front>
        <title>Extensible Markup Language (XML) 1.0 (Fifth Edition)</title>
        <author initials="T." surname="Bray" fullname="Tim Bray" role="editor"/>
        <author initials="J." surname="Paoli" fullname="Jean Paoli" role="editor"/>
        <author initials="C.M." surname="Sperberg-McQueen" fullname="C. M. Sperberg McQueen" role="editor"/>
        <author initials="E." surname="Maler" fullname="Eve Maler" role="editor"/>
        <author initials="F." surname="Yergeau" fullname="Francois Yergeau" role="editor"/>
        <date day="26" month="November" year="2008"/>
      </front>
      <refcontent>W3C Recommendation</refcontent>
    </reference>

    <reference anchor="YAML">
        <front>
            <title>Yet Another Markup Language (YAML) 1.0</title>
            <author initials="B." surname="Ingerson" />
            <author initials="C." surname="Evans" />
            <author initials="O." surname="Ben-Kiki" />
            <date month="October" year="2023" />
        </front>
        <seriesInfo name="Available:" value="https://yaml.org/spec/history/2001-05-26.html" />
    </reference>



    <reference anchor="TS-23.501">
        <front>
            <title>System Architecture for the 5G System (5GS)</title>
            <author surname="3GPP TS 23.501 V18.3.0" />
            <date month="September" year="2023" />
        </front>
        <seriesInfo name="Available:" value="https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3144" />
    </reference>

    <reference anchor="TS-28.312">
        <front>
            <title>Intent Driven Management Services for Mobile Networks</title>
            <author surname="3GPP TS 28.312 V18.1.1" />
            <date month="September" year="2023" />
        </front>
        <seriesInfo name="Available:" value="https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3554" />
    </reference>

    <reference anchor="TR-28.812">
        <front>
            <title>Study on Scenarios for Intent Driven Management Services for Mobile Networks</title>
            <author surname="3GPP TR 28.812 V17.1.0" />
            <date month="December" year="2020" />
        </front>
        <seriesInfo name="Available:" value="https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3553" />
    </reference>

    <reference anchor="TS-23.288">
        <front>
            <title>Architecture Enhancements for 5G System (5GS) to Support Network Data Analytics Services</title>
            <author surname="3GPP TS 23.288 V18.3.0" />
            <date month="September" year="2023" />
        </front>
        <seriesInfo name="Available:" value="https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3579" />
    </reference>

    <reference anchor="TS-29.520">
        <front>
            <title>Network Data Analytics Services</title>
            <author surname="3GPP TS 29.520 V18.3.0" />
            <date month="September" year="2023" />
        </front>
        <seriesInfo name="Available:" value="https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3355" />
    </reference>

    <reference anchor="ETSI-NFV">
        <front>
            <title>Network Functions Virtualisation (NFV); Architectural Framework</title>
            <author surname="ETSI GS NFV 002 V1.2.1" />
            <date month="December" year="2014" />
        </front>
        <seriesInfo name="Available:" value="https://www.etsi.org/deliver/etsi_gs/nfv/001_099/002/01.02.01_60/gs_nfv002v010201p.pdf" />
    </reference>

    <reference anchor="ETSI-NFV-Release-2">
        <front>
            <title>Network Functions Virtualisation (NFV) Release 2; 
            Management and Orchestration; Architectural Framework Specification</title>
            <author surname="ETSI GS NFV 006 V2.1.1" />
            <date month="January" year="2021" />
        </front>
        <seriesInfo name="Available:" value="https://www.etsi.org/deliver/etsi_gs/nfv/001_099/006/02.01.01_60/gs_nfv006v020101p.pdf" />
    </reference>

    <reference anchor="REST">
        <front>
            <title>Principled Design of the Modern Web Architecture</title>
            <author initials="R." surname="Fielding" />
            <author initials="R." surname="Taylor" />
            <date month="May" year="2002" />
        </front>
        <seriesInfo name="ACM" value="Transactions on Internet Technology, Vol. 2, Issue 2" />
        <seriesInfo name="Available:" value="https://dl.acm.org/doi/10.1145/514183.514185" />
    </reference>

    <reference anchor="GPT-3">
        <front>
            <title>Language Models are Few-Shot Learners</title>
            <author initials="T." surname="Brown" />
            <date month="May" year="2020" />
        </front>
        <seriesInfo name="" value="arXiv:2005.14165" />
        <seriesInfo name="Available:" value="https://arxiv.org/abs/2005.14165" />
    </reference>

    <reference anchor="Flan-T5">
        <front>
            <title>Scaling Instruction-Finetuned Language Models</title>
            <author initials="H." surname="Chung" />
            <date month="October" year="2022" />
        </front>
        <seriesInfo name="" value="arXiv:2210.11416" />
        <seriesInfo name="Available:" value="https://arxiv.org/abs/2210.11416" />
    </reference>            

    <reference anchor="USENIX-ATC-Lumi">
        <front>
            <title>Hey, Lumi! Using Natural Language for Intent-Based Network Management</title>
            <author initials="A." surname="Jacobs" />
            <author initials="R." surname="Pfitscher" />
            <author initials="R." surname="Ribeiro" />
            <author initials="R." surname="Ferreira" />
            <author initials="L." surname="Granville" />
            <author initials="W." surname="Willinger" />
            <author initials="S." surname="Rao" />
            <date month="July" year="2021" />
        </front>
        <seriesInfo name="USENIX" value="Annual Technical Conference" />
        <seriesInfo name="Available:" value="https://www.usenix.org/conference/atc21/presentation/jacobs" />
    </reference>

    <reference anchor="BERT">
        <front>
            <title>BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding</title>
            <author initials="J." surname="Devlin" />
            <author initials="M." surname="Chang" />
            <author initials="K." surname="Lee" />
            <author initials="K." surname="Toutanova" />
            <date month="June" year="2019" />
        </front>
        <seriesInfo name="NAACL-HLT" value="Conference" />
        <seriesInfo name="Available:" value="https://aclanthology.org/N19-1423.pdf" />
    </reference>

    <reference anchor="AUTOSAR-SDV">
        <front>
            <title>AUTOSAR Adaptive Platform</title>
            <author surname="AUTOSAR" />
            <date month="March" year="2024" />
        </front>
        <seriesInfo name="Available:" value="https://www.autosar.org/standards/adaptive-platform" />    
    </reference>

    <reference anchor="Eclipse-SDV">
        <front>
            <title>Eclipse Software Defined Vehicle Working Group Charter</title>
            <author surname="Eclipse" />
            <date month="March" year="2024" />
        </front>
        <seriesInfo name="Available:" value="https://www.eclipse.org/org/workinggroups/sdv-charter.php" />    
    </reference>

    <reference anchor="COVESA">
        <front>
            <title>Connected Vehicle Systems Alliance </title>
            <author surname="COVESA" />
            <date month="March" year="2024" />
        </front>
        <seriesInfo name="Available:" value="https://covesa.global/" />    
    </reference>

    <reference anchor="A2A">
        <front>
            <title>Agent2Agent (A2A) Protocol</title>
            <author surname="Linux Foundation" />
            <date month="November" year="2026" />
        </front>
        <seriesInfo name="Available:" value="https://a2a-protocol.org/" />    
    </reference>

    <reference anchor="A2A-Specification">
        <front>
            <title>Agent2Agent (A2A) Protocol Specification</title>
            <author surname="Linux Foundation" />
            <date month="November" year="2026" />
        </front>
        <seriesInfo name="Available:" value="https://a2a-protocol.org/latest/specification/" />    
    </reference>

    <reference anchor="MCP">
        <front>
            <title>Model Context Protocol (MCP)</title>
            <author surname="Linux Foundation" />
            <date month="November" year="2026" />
        </front>
        <seriesInfo name="Available:" value="https://modelcontextprotocol.io/" />    
    </reference>

    <reference anchor="MCP-Specification">
        <front>
            <title>Model Context Protocol Specification</title>
            <author surname="Linux Foundation" />
            <date month="November" year="2026" />
        </front>
        <seriesInfo name="Available:" value="https://modelcontextprotocol.info/specification/" />    
    </reference>

    <reference anchor="Kubernetes">
        <front>
            <title>Kubernetes: Cloud Native Computing Platform</title>
            <author surname="Kubernetes" />
            <date month="March" year="2024" />
        </front>
        <seriesInfo name="Available:" value="https://kubernetes.io/" />    
    </reference>

    <reference anchor="Survey-IBN-CST-2023">
        <front>
            <title>A Survey on Intent-Based Networking</title>
            <author initials="A." surname="Leivadeas" />
            <author initials="M." surname="Falkner" />
            <date month="March" year="2023" />
        </front>
        <seriesInfo name="Available:" value="https://ieeexplore.ieee.org/document/9925251" />    
    </reference>

    <reference anchor="ClickINC">
        <front>
            <title>ClickINC: In-network Computing as a Service in Heterogeneous Programmable Data-center Networks</title>
            <author initials="W." surname="Xu" />
            <author initials="Z." surname="Zhang" />
            <author initials="Y." surname="Feng" />
            <author initials="H." surname="Song" />
            <author initials="Z." surname="Chen" />
            <author initials="W." surname="Wu" />
            <author initials="G." surname="Liu" />
            <author initials="Y." surname="Zhang" />
            <author initials="S." surname="Liu" />
            <author initials="Z." surname="Tian" />
            <author initials="B." surname="Liu" />
            <date month="September" year="2023" />
        </front>
        <seriesInfo name="Publisher:" value="ACM SIGCOMM" />
        <seriesInfo name="Available:" value="https://dl.acm.org/doi/10.1145/3603269.3604835" />    
    </reference>

</references>
<!-- END: Informative References -->

<!-- START: Changes -->
<section title="Changes from draft-jeong-nmrg-agentic-ai-framework-00">
    <t>
      The following changes are made from draft-jeong-nmrg-agentic-ai-framework-00:
      <ul>
        <li>
        This version has corrected the Agentic AI System in
        <xref target="figure:Agentic-AI-System-for-Service-Management" />
        with Wired/Wireless Link.
        </li>
        <li>
        This version has clarified the Agentic AI Framework in
        <xref target="figure:Agentic-AI-Framework-for-Service-Management" />
        with detailed descriptions on the framework's components and procedures.
        </li>
      </ul>
    </t>
</section>
<!-- END: Changes -->

<!-- START: Acknowledgments -->
<section anchor="section:Acknowledgments" numbered="false" title="Acknowledgments">
    <t>
    This document is a product of the NMRG including the RG Chairs 
    (Jerome Francois and Jeferson Campos Nobre).  This document benefitted
    from the review and comments from the following people: 
    Yong-Geun Hong and Joo-Sang Youn.
    The authors sincerely appreciate their efforts and kind help.
    </t>

    <t>    
    This work was supported by the Institute of Information &amp; Communications
    Technology Planning &amp; Evaluation (IITP) grant funded by the Korea
    Ministry of Science and ICT (MSIT) (No. RS-2024-00398199).
    </t>
</section>
<!-- END: Acknowledgments -->

<!-- START: Contributors -->
<section anchor="section:Contributors" numbered="false" title="Contributors">
    <t>  
    The following are coauthors of this document:
    </t>   

      <contact fullname="Jisuk Chae">
        <organization showOnFrontPage="true">Department of Computer Science &amp; Engineering</organization>
        <address>
          <postal>
            <extaddr>Sungkyunkwan University</extaddr>
            <street>2066 Seobu-Ro, Jangan-Gu</street>
            <city>Suwon</city>
            <region>Gyeonggi-Do</region>
            <code>16419</code>
            <country>Republic of Korea</country>
          </postal>
          <phone>+82 31 299 4106</phone>
          <email>sue030124@skku.edu</email>
          <uri>http://iotlab.skku.edu/people-Jisuk-Chae.php</uri>
        </address>
      </contact>
      <contact fullname="Eunjin Hwang">
        <organization showOnFrontPage="true">Department of Computer Science &amp; Engineering</organization>
        <address>
          <postal>
            <extaddr>Sungkyunkwan University</extaddr>
            <street>2066 Seobu-Ro, Jangan-Gu</street>
            <city>Suwon</city>
            <region>Gyeonggi-Do</region>
            <code>16419</code>
            <country>Republic of Korea</country>
          </postal>
          <phone>+82 31 299 4106</phone>
          <email>eunijhwang@skku.edu</email>
          <uri>https://iotlab.skku.edu/people-Eunjin-Hwang.php</uri>
        </address>
      </contact>
      <contact fullname="KiHyun Shim">
        <organization showOnFrontPage="true">Department of Computer Science &amp; Engineering</organization>
        <address>
          <postal>
            <extaddr>Sungkyunkwan University</extaddr>
            <street>2066 Seobu-Ro, Jangan-Gu</street>
            <city>Suwon</city>
            <region>Gyeonggi-Do</region>
            <code>16419</code>
            <country>Republic of Korea</country>
          </postal>
          <phone>+82 31 299 4106</phone>
          <email>tlarlguswh@skku.edu</email>
          <uri>https://iotlab.skku.edu/people-KiHyun_Shim.php</uri>
        </address>
      </contact>                  
</section>
<!-- END: Contributors -->

</back>

<!-- <vspace blankLines="100"/> -->
<!-- page break to put addresses onto one page-->

</rfc>
