<?xml version='1.0' encoding='utf-8'?>
<!DOCTYPE rfc [
  <!ENTITY nbsp    "&#160;">
  <!ENTITY zwsp   "&#8203;">
  <!ENTITY nbhy   "&#8209;">
  <!ENTITY wj     "&#8288;">
]>
<?xml-stylesheet type="text/xsl" href="rfc2629.xslt" ?>
<!-- generated by https://github.com/cabo/kramdown-rfc version 1.7.43 (Ruby 3.2.3) -->
<rfc xmlns:xi="http://www.w3.org/2001/XInclude" ipr="trust200902" docName="draft-ietf-agentproto-usecases-requirements-00" category="info" submissionType="IETF" tocInclude="true" sortRefs="true" symRefs="true" version="3">
  <!-- xml2rfc v2v3 conversion 3.35.0 -->
  <front>
    <title abbrev="agentic-ai-ucreq">Agentic AI Use Cases and Requirements</title>
    <seriesInfo name="Internet-Draft" value="draft-ietf-agentproto-usecases-requirements-00"/>
    <author initials="T." surname="Reddy" fullname="Tirumaleswar Reddy">
      <organization>Nokia</organization>
      <address>
        <postal>
          <city>Bangalore</city>
          <region>Karnataka</region>
          <country>India</country>
        </postal>
        <email>kondtir@gmail.com</email>
      </address>
    </author>
    <author initials="Z." surname="Sarker" fullname="Zaheduzzaman Sarker">
      <organization>Nokia</organization>
      <address>
        <postal>
          <country>Sweden</country>
        </postal>
        <email>zaheduzzaman.sarker@nokia.com</email>
      </address>
    </author>
    <author initials="K." surname="Yao" fullname="Kehan Yao">
      <organization>China Mobile</organization>
      <address>
        <email>yaokehan@chinamobile.com</email>
      </address>
    </author>
    <date year="2026" month="November" day="20"/>
    <area>Applications and Real-Time</area>
    <workgroup>Agent Communication Protocols</workgroup>
    <keyword>Internet-Draft</keyword>
    <abstract>
      <?line 88?>

<t>This document describes use cases for agentic AI communication systems
and derives protocol requirements from those use cases. The requirements
are intended to guide IETF standardization work on protocols in the
context of agent-to-agent communication, agent-to-tool communication,
with focus on dialog management, transport, multimodal communication,
communication security, agent identity and authentication.</t>
    </abstract>
  </front>
  <middle>
    <?line 97?>

<section anchor="introduction">
      <name>Introduction</name>
      <t>An AI agent is an autonomous, adaptive intelligent software system
that uses AI models to complete a specific objective on behalf of
a user or another AI agent. It makes decisions, executes actions,
and interacts with other agents through tasks and tools. Unlike
traditional software workloads that follow fixed execution paths, an
AI agent dynamically determines at run time which actions to take,
which tools to invoke, and which agents to collaborate with, based on
reasoning over its goals and context.</t>
      <t>This document presents use cases that illustrate the key interaction
patterns of agentic AI communication systems, and derives protocol
requirements from those use cases. The requirements are intended to
drive development of the agentic dialog management protocol, its
transport bindings, and the reference architecture for agentic AI systems.</t>
      <t>Each use case contributes a distinct slice of requirements, and it is their
composition that distinguishes an agentic communication protocol from ordinary
application-to-service invocation.</t>
      <t>The use cases in this document cover interaction patterns for
agentic AI communication systems. This document takes into account
related use case and problem statement documents including <xref target="SCRM"/>,
<xref target="YAO"/>, and <xref target="SONG"/>, and existing protocol work including
<xref target="A2A"/> and <xref target="MCP"/>.</t>
    </section>
    <section anchor="terminology">
      <name>Terminology</name>
      <t><strong>AI Agent</strong>: An autonomous software entity that perceives its
environment, maintains internal state, and executes actions to achieve
specified goals, potentially including communication with other agents
or invocation of external tools.</t>
      <t><strong>Agent Identity</strong>: Identity information associated with an AI agent,
used for authentication and accountability within agentic communication
systems.</t>
      <t><strong>Agentic AI Communication System</strong>: A system comprising one or more
AI agents that communicate with each other, with users, and with
external tools or services to complete tasks. The communication
interfaces between these entities are the subject of protocol
standardization in this document.</t>
      <t><strong>Agent-to-Agent Communication</strong>: Direct or brokered
communication between two or more AI agents, where brokered
communication involves an intermediary agent or coordination service,
as distinguished from communication between an agent and a user or
between an agent and a tool.</t>
      <t><strong>Capability</strong>: A description of what an agent can
perform, including inputs, outputs, constraints, and required conditions.</t>
      <t><strong>Coordinator Agent</strong>: An agent that distributes a shared problem or task
to a group of peer agents, aggregates their outputs, and iteratively drives
them toward a collective result or consensus.</t>
      <t><strong>Delegation</strong>: The act of an agent requesting another agent to execute
a task on its behalf.</t>
      <t><strong>Initiating Agent</strong>: An agent that receives an initial request and
delegates subtasks to peer agents. Any peer agent may itself delegate
further to other agents without routing through the initiating agent.</t>
      <t><strong>Mediator</strong>: An entity that serves as a proxy or mediator for
external tools, APIs, databases, or other resources that other agents
require but cannot directly access.</t>
      <t><strong>Message</strong>: A discrete unit of communication exchanged between agents
containing structured data such as a task request, response, progress
update, event notification, or control signal.</t>
      <t><strong>Modality</strong>: A category of data format used for input or output in
agent communication, such as text, audio, image, or video. A dialog
may support one or more modalities simultaneously.</t>
      <t><strong>Orchestrator Agent</strong>: An agent that acts as a controller,
coordinating the activity of other agents by decomposing goals into
sub-tasks and delegating those sub-tasks to appropriate agents.</t>
      <t><strong>Peer Agent</strong>: An agent that receives delegated subtasks from another
agent and may itself delegate further to other agents.</t>
      <t><strong>Dialog</strong>: The correlated sequence of interactions between two or
more participants (users, agents, and tools) over the course of a task. A
dialog may persist across multiple individual message exchanges and network
connections.</t>
      <t><strong>Dialog Context</strong>: The identifiers and lifecycle state needed to
correlate and maintain a dialog. It does not include the content exchanged
with AI models, such as conversation memory, retrieved documents, or prompts.</t>
      <t><strong>Linked Dialog</strong>: A new dialog, with a new dialog context, that a
participant creates to continue a dialog it received.</t>
      <t><strong>Task Context</strong>: The data that a participant maintains
to execute a task, such as conversation history and intermediate
results. Task context is not carried by the protocol.</t>
      <t><strong>Task</strong>: A unit of work submitted by a user to an agent, or
delegated by one agent to another.</t>
      <t><strong>Task State</strong>: The current execution status of a task (e.g., pending,
in-progress, completed, failed).</t>
      <t><strong>Tool</strong>: An external service invoked by an agent to retrieve data or
perform operations. A tool is not necessarily an agent and may not participate
in agent-to-agent communication.</t>
      <t><strong>User</strong>: A human that initiates interaction with an
AI agent by submitting a request or task.</t>
      <section anchor="dialog-task-context">
        <name>Dialog Context and Task Context</name>
        <t>Any participant, including an intermediary, can use
dialog context to correlate and maintain a dialog without interpreting
the task. For example, if the transport connection between two agents
is interrupted, the dialog identifier allows the reconnecting agent to
re-associate with the existing dialog instead of starting a new dialog
and restarting the task. An intermediary uses the dialog identifier to
correlate the requests it receives with the requests it forwards.
Neither function requires access to the task context. Each participant
can use the dialog identifier to retrieve its own task context.</t>
      </section>
    </section>
    <section anchor="common-requirements">
      <name>Common Protocol Requirements</name>
      <t>The following baseline requirements apply to both agent-to-agent and agent-to-tool protocol interactions across all use cases and are not repeated per use case.</t>
      <t>Each per-use-case requirement is tagged with one or more of the following protocol area tags to allow cross-use-case navigation:</t>
      <ul spacing="normal">
        <li>
          <t><strong>Dialog</strong>: Requirements related to dialog correlation, lifecycle, continuity, and recovery.</t>
        </li>
        <li>
          <t><strong>Transport</strong>: Requirements related to message delivery, streaming, cancellation, and data transfer.</t>
        </li>
        <li>
          <t><strong>Security</strong>: Requirements related to confidentiality, integrity, and authorization.</t>
        </li>
        <li>
          <t><strong>Authentication</strong>: Requirements related to identity verification and credential delegation.</t>
        </li>
      </ul>
      <table>
        <thead>
          <tr>
            <th align="left">REQ-ID</th>
            <th align="left">Description</th>
            <th align="left">Tag</th>
          </tr>
        </thead>
        <tbody>
          <tr>
            <td align="left">CMN-1</td>
            <td align="left">The protocol is required to support any client application to communicate with any agent service.</td>
            <td align="left">Authentication</td>
          </tr>
          <tr>
            <td align="left">CMN-2</td>
            <td align="left">Mutual authentication is required between all communicating parties.</td>
            <td align="left">Authentication</td>
          </tr>
          <tr>
            <td align="left">CMN-3</td>
            <td align="left">All protocol traffic is required to be encrypted and integrity-protected in transit.</td>
            <td align="left">Security</td>
          </tr>
          <tr>
            <td align="left">CMN-4</td>
            <td align="left">Structured error responses are required to include an authorization scope violation type, reported by the orchestrator or mediator when an agent attempts an action that exceeds or contradicts the scope delegated to it.</td>
            <td align="left">Security</td>
          </tr>
          <tr>
            <td align="left">CMN-5</td>
            <td align="left">Structured error responses are required, distinguishing at minimum: authentication failure, authorization failure, timeout, and internal error.</td>
            <td align="left">Transport</td>
          </tr>
          <tr>
            <td align="left">CMN-6</td>
            <td align="left">The protocol is required to provide a means to signal task priority so that critical-path tasks can be scheduled ahead of lower-priority ones.</td>
            <td align="left">Transport</td>
          </tr>
          <tr>
            <td align="left">CMN-7</td>
            <td align="left">The protocol is required to support cryptographic algorithm agility, ensuring that cryptographic algorithms used for encryption, authentication, credential verification, and integrity protection can be negotiated and updated over time, in accordance with <xref target="RFC7696"/>.</td>
            <td align="left">Security, Authentication</td>
          </tr>
          <tr>
            <td align="left">CMN-8</td>
            <td align="left">The protocol is required to provide a means to verify the agent authentication credentials validity used by agents at the time of use.</td>
            <td align="left">Authentication</td>
          </tr>
          <tr>
            <td align="left">CMN-9</td>
            <td align="left">The protocol is required to support signaling that a presented credential has been revoked or is otherwise invalid.</td>
            <td align="left">Security</td>
          </tr>
          <tr>
            <td align="left">CMN-10</td>
            <td align="left">The protocol is required to support revoking a previously granted authorization, including propagating the revocation across a delegation chain so that affected agents cease to act under it.</td>
            <td align="left">Security</td>
          </tr>
          <tr>
            <td align="left">CMN-11</td>
            <td align="left">The protocol is required to support establishment, modification, and termination of a dialog, distinct from cancellation of a delegated subtask within the dialog. Modification includes changes to the modalities or lifetime of the dialog. The protocol is required to support a dialog lifetime, after which the dialog is terminated.</td>
            <td align="left">Dialog</td>
          </tr>
          <tr>
            <td align="left">CMN-12</td>
            <td align="left">Before accepting a change to a dialog, a participant is required to be able to verify that the change is authorized at that time. If this check fails or cannot be completed, the participant is required to reject the change and signal the reason. It is required to be able to determine which of the two cases occurred.</td>
            <td align="left">Dialog, Security</td>
          </tr>
        </tbody>
      </table>
      <section anchor="network-assumptions">
        <name>Network-Layer Assumptions</name>
        <t>The requirements in this document assume that messages
that establish, modify, resume, or terminate a dialog are delivered
reliably and in order. Data for real-time modalities, such as
interactive audio and video, can be delivered without reliability or
ordering to meet latency requirements. All the deliveries are
congestion controlled. Whether these share a transport connection is
determined by the transport bindings.</t>
      </section>
    </section>
    <section anchor="usecases">
      <name>Use Cases</name>
      <t>These interactions differ from ordinary application-to-service invocation in
several respects, one being runtime selection of agents and tools by capability.
An agent's internal processing, including perception, planning and
re-planning, and invocation of its AI model, is not visible on the protocol
interface and is out of scope; it motivates the use cases but drives no
protocol requirement. Cross-domain operation can apply to any of
these use cases. Crossing an administrative boundary adds authentication and
authorization requirements, which map to the OAuth and WIMSE work discussed in
<xref target="relationship-oauth-wimse"/>, and privacy requirements on dialog context
propagated across the boundary, discussed in <xref target="security"/>. The use
cases are independent of deployment model: they apply whether participants are
within a single administrative domain or span several, and whether they
interact directly or through intermediaries. Together they cover the
deployment models in scope: a client application on a user's device and
an agent (<xref target="simple-single-agent"/>), agents hosted by different
organizations or service providers (<xref target="peer-collaborative"/>), and agents
reached through a mediator such as a network operator's agent gateway
(<xref target="tool-mediation"/>).</t>
      <section anchor="simple-single-agent">
        <name>Simple Single-Agent Task</name>
        <section anchor="description">
          <name>Description</name>
          <t>A user submits a task to an AI agent via a client application. The
client application initiates a dialog with the agent, which creates and
maintains it. The agent executes the task by invoking one or more tools and returns
results to the user. The tools invoked by the agent may reside in
the same or a different administrative domain. The agent protocol
is required to support multiple input and output modalities, and
the client application and agent are required to be able to negotiate which
modalities are active for the dialog.</t>
          <t>This use case covers the protocol interface between the client
application and the agent. The interaction between the user and
the client application is out of scope in this use case. This use case assumes that
the user communicates with the agent via a client application;
direct communication between a user and an agent without an
intermediary client application is not covered in this use case.</t>
          <t>This interaction pattern is described in <xref target="SCRM"/>.</t>
        </section>
        <section anchor="interaction-flow">
          <name>Interaction Flow</name>
          <artwork><![CDATA[
+----------------+                       +-----------+
| App/agent      |<--------------------> |   Agent   |
+----------------+        Protocol       +-----------+
                                               |
                                     Protocol  |
                                               v
                                          +---------+
                                          | Tool(s) |
                                          +---------+
]]></artwork>
        </section>
        <section anchor="example">
          <name>Example</name>
          <t>A user books a flight through a travel booking app. The app is the
client application. It sends the user's request to a travel agent run
by the travel company that provides the app. The agent uses two
tools: a fare search tool run by the travel company, and a booking
tool run by an airline, in a different administrative domain.</t>
          <artwork><![CDATA[
+------------+      +--------------+      +----------------+
| Travel app |<---->| Travel agent |----->| Fare search    |
|            |      |              |      | (travel co.)   |
+------------+      +--------------+      +----------------+
                           |
                           |              +----------------+
                           +------------->| Flight booking |
                                          | (airline)      |
                                          +----------------+
]]></artwork>
          <ol spacing="normal" type="1"><li>
              <t>The app establishes a dialog with the agent and asks for a flight on
a given route and date (CMN-11).</t>
            </li>
            <li>
              <t>The agent queries the fare search tool and returns matching flights
to the app.</t>
            </li>
            <li>
              <t>The user picks a flight. The agent calls the airline's booking
tool with a credential showing that the user authorized the
booking, without the user's airline credentials (A1-9).</t>
            </li>
            <li>
              <t>The agent returns the booking confirmation, and the app terminates
the dialog (CMN-11).</t>
            </li>
          </ol>
          <t>If the user abandons a search before step 3, the app cancels that task
(A1-2). The dialog continues, and the user can start a new search in
it without starting over.</t>
        </section>
        <section anchor="a1-protocol-requirements">
          <name>Protocol Requirements</name>
          <t>This use case involves two distinct protocol interfaces. The App/Agent-to-Agent
interface is between the party making the request, which may be a client
application or another agent, and the agent that handles it. The Agent-to-Tool
interface is between the agent and the tools it invokes. A given requirement
does not necessarily apply to both; the Interaction column indicates which
interface each applies to.</t>
          <table>
            <thead>
              <tr>
                <th align="left">REQ-ID</th>
                <th align="left">Description</th>
                <th align="left">Interaction</th>
                <th align="left">Tag</th>
              </tr>
            </thead>
            <tbody>
              <tr>
                <td align="left">A1-1</td>
                <td align="left">The protocol is required to support incremental streaming of agent output, allowing partial results to be delivered to the client before the agent has completed processing.</td>
                <td align="left">App/Agent-to-Agent</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">A1-2</td>
                <td align="left">The protocol is required to define a task cancellation message that the client can issue at any point during task execution.</td>
                <td align="left">App/Agent-to-Agent</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">A1-3</td>
                <td align="left">The protocol is required to define structured error message types that distinguish at minimum: transport failure, tool invocation failure, and agent processing failure.</td>
                <td align="left">Both</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">A1-4</td>
                <td align="left">The protocol is required to support multiple modalities for both input and output, including real-time audio and video with low latency.</td>
                <td align="left">App/Agent-to-Agent</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">A1-5</td>
                <td align="left">The protocol is required to support modality negotiation at dialog establishment, allowing the client and agent to agree on which modalities are active for the dialog.</td>
                <td align="left">App/Agent-to-Agent</td>
                <td align="left">Dialog, Transport</td>
              </tr>
              <tr>
                <td align="left">A1-6</td>
                <td align="left">The protocol is required to support agent-initiated notifications to the client during task execution.</td>
                <td align="left">App/Agent-to-Agent</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">A1-7</td>
                <td align="left">The protocol is required to support concurrent invocation of multiple tools within a single agent task, where tools may be operated by distinct providers across different administrative domains, each with independent authentication and authorization requirements.</td>
                <td align="left">Agent-to-Tool</td>
                <td align="left">Transport, Security</td>
              </tr>
              <tr>
                <td align="left">A1-8</td>
                <td align="left">The protocol is required to support bulk transfer of large data between communicating parties, applicable to both agent-to-tool and agent-to-agent interactions.</td>
                <td align="left">Both</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">A1-9</td>
                <td align="left">A delegation mechanism is required to be defined by which an agent presents to a tool provider a credential attesting the authorization for the requested tool access, without exposing the client's primary credentials.</td>
                <td align="left">Agent-to-Tool</td>
                <td align="left">Authentication</td>
              </tr>
              <tr>
                <td align="left">A1-10</td>
                <td align="left">The protocol is required to preserve the dialog context when a participant transforms data from one modality to another, such as audio to text.</td>
                <td align="left">App/Agent-to-Agent</td>
                <td align="left">Dialog</td>
              </tr>
              <tr>
                <td align="left">A1-11</td>
                <td align="left">The protocol is required to allow the user to continue a dialog from the same client application on another device.</td>
                <td align="left">App/Agent-to-Agent</td>
                <td align="left">Dialog</td>
              </tr>
            </tbody>
          </table>
        </section>
      </section>
      <section anchor="orchestrator-agent">
        <name>Orchestrator and Agent Collaboration</name>
        <section anchor="description-1">
          <name>Description</name>
          <t>An orchestrator agent acts as a controller, decomposing a task into
subtasks and delegating them asynchronously to one or more other agents.
The orchestrator decides which other agents to invoke, sequences the
delegation, and aggregates results to continue task execution. Each
agent executes the respective subtask independently and reports results back
to the orchestrator.</t>
          <t>It should be noted that AI models are stateless by nature: each inference
call processes only what is explicitly provided with a particular context,
with no persistent memory between calls. The application
layer is responsible for maintaining the context across the calls by
carrying conversation history, intermediate results, and task
state. Dialog continuity is therefore required to preserve the
dialog context across network interruptions, so that a reconnecting
agent can re-associate with the dialog and use it to restore its own
task context without reconstructing it from scratch.</t>
          <t>If a dialog is interrupted before the result of an operation reaches
the delegating agent, the operation may or may not have completed.
Resuming the dialog does not tell the delegating agent which. Before
resending the operation, the delegating agent checks its outcome with
the delegated agent or tool that performed it. If the outcome cannot be
determined, the operation remains unresolved, and resuming the dialog
alone does not justify resending it.</t>
          <t>This pattern is reflected in <xref target="A2A"/>,
and is implemented in deployed multi-agent frameworks including
<xref target="AUTOGEN"/>, <xref target="LANGCHAIN"/>, and <xref target="OPENAI-AGENTS"/>.</t>
        </section>
        <section anchor="interaction-flow-1">
          <name>Interaction Flow</name>
          <artwork><![CDATA[
+---------------------+                        +------------+
| Orchestrator Agent  |---Task Delegation----->|   Agent-1  |
|                     |<--Result Reporting-----|            |
|                     |                        +------------+
|                     |
|                     |                        +------------+
|                     |---Task Delegation----->|   Agent-2  |
|                     |<--Result Reporting-----|            |
+---------------------+                        +------------+
]]></artwork>
        </section>
        <section anchor="example-1">
          <name>Example</name>
          <t>A user asks a trip planning agent, the orchestrator, to arrange a
business trip. The orchestrator delegates the flight booking to a
flight agent and the hotel booking to a hotel agent. Both agents are
run by different providers.</t>
          <artwork><![CDATA[
+--------------+      +--------------+
|              |----->| Flight agent |
| Trip planner |      +--------------+
| (orchestr.)  |
|              |      +--------------+
|              |----->| Hotel agent  |
+--------------+      +--------------+
]]></artwork>
          <ol spacing="normal" type="1"><li>
              <t>The orchestrator establishes a dialog with each agent and delegates
both subtasks without waiting for either to finish (B1-1, B1-2).</t>
            </li>
            <li>
              <t>The flight agent streams fare options and reports the booked
flight (B1-4, B1-3).</t>
            </li>
            <li>
              <t>The orchestrator sends the hotel agent a request to book a room.
The hotel agent books the room, but the connection drops before its
confirmation reaches the orchestrator.</t>
            </li>
            <li>
              <t>The orchestrator reconnects and resumes the dialog with the hotel
agent (B1-6). The hotel agent verifies that the orchestrator is
the participant whose exchange was interrupted, and that the dialog
is current and still authorized (B1-7).</t>
            </li>
            <li>
              <t>Instead of sending the booking request again, which could book a
second room, the orchestrator asks the hotel agent for the outcome
of its request. It learns that the room was booked and uses that
booking.</t>
            </li>
          </ol>
        </section>
        <section anchor="b1-protocol-requirements">
          <name>Protocol Requirements</name>
          <table>
            <thead>
              <tr>
                <th align="left">REQ-ID</th>
                <th align="left">Description</th>
                <th align="left">Tag</th>
              </tr>
            </thead>
            <tbody>
              <tr>
                <td align="left">B1-1</td>
                <td align="left">The protocol is required to facilitate task delegation for an orchestrator to agents that includes task delegation and acknowledgement message types.</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">B1-2</td>
                <td align="left">The protocol is required to support asynchronous delegation, allowing the orchestrator to delegate to multiple agents without waiting for each to complete before proceeding.</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">B1-3</td>
                <td align="left">The protocol is required to define a result reporting message by which an agent returns its completed output to the orchestrator.</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">B1-4</td>
                <td align="left">The protocol is required to support streaming of intermediate results from the agent to the orchestrator during task execution.</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">B1-5</td>
                <td align="left">The protocol is required to define a task cancellation message that the orchestrator can send to an agent to abort a delegated subtask.</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">B1-6</td>
                <td align="left">The protocol is required to support persistent dialog identifiers that survive network interruption or change, and is required to define a dialog resumption message by which an agent re-attaches to an interrupted dialog, restoring the prior dialog context.</td>
                <td align="left">Dialog</td>
              </tr>
              <tr>
                <td align="left">B1-7</td>
                <td align="left">On resumption, the protocol is required to enable the receiving participant to verify that (1) the party resuming the dialog is the participant whose exchange was interrupted, (2) the dialog context presented is the current state of that dialog, and (3) the dialog has not been terminated and its authorization has not been revoked.</td>
                <td align="left">Dialog, Security</td>
              </tr>
              <tr>
                <td align="left">B1-8</td>
                <td align="left">If any check in B1-7 fails or cannot be completed, the dialog is not resumed and the protocol is required to signal the reason. The receiving participant is required to be able to determine whether the check failed or could not be completed.</td>
                <td align="left">Dialog, Security</td>
              </tr>
            </tbody>
          </table>
        </section>
      </section>
      <section anchor="authz-checkpoint">
        <name>Long-Running Delegated Task with Authorization Checkpoint</name>
        <section anchor="description-2">
          <name>Description</name>
          <t>An orchestrator agent delegates a long-running task to other agents.
The delegated agent executes the task autonomously and sends progress
notifications to the orchestrator. At any certain point one or more delegated
agents pause and request explicit authorization from the orchestrator
before proceeding further. The orchestrator may relay this authorization
request to the invoker (user or agent) or resolve it autonomously based on policy.</t>
          <t>This pattern is reflected in the In-Task Authorization mechanism
defined in <xref target="A2A"/>.</t>
        </section>
        <section anchor="interaction-flow-2">
          <name>Interaction Flow</name>
          <artwork><![CDATA[
+---------------------+                        +------------+
| Orchestrator Agent  |---Task Delegation----->|            |
|                     |<--Progress Notif.------|            |
|                     |<--Authz Checkpoint-----|  Agent(s)  |
|                     |---Authz Response------>|            |
|                     |<--Result Reporting-----|            |
+---------------------+                        +------------+
]]></artwork>
        </section>
        <section anchor="example-2">
          <name>Example</name>
          <t>A user asks a banking assistant, the orchestrator, to close an
account. The orchestrator delegates the closure to an account closure
agent, which runs a workflow that can take several days.</t>
          <artwork><![CDATA[
+------------+      +--------------+      +---------------+
|    User    |<---->|   Banking    |<---->| Account       |
|            |      |  assistant   |      | closure agent |
+------------+      +--------------+      +---------------+
]]></artwork>
          <ol spacing="normal" type="1"><li>
              <t>The orchestrator establishes a dialog with the closure agent and
delegates the closure. The dialog lasts for the whole workflow.</t>
            </li>
            <li>
              <t>The closure agent cancels the account's standing payments and
closes the linked debit card. After each step it sends the
orchestrator a progress notification, such as "standing payments
cancelled" (B2-1).</t>
            </li>
            <li>
              <t>Before transferring the remaining balance, the closure agent
pauses and sends an authorization checkpoint request stating the
amount and the destination account (B2-2, B2-6).</t>
            </li>
            <li>
              <t>The orchestrator relays the request to the user, who approves it
(B2-3). The approval covers only this transfer (B2-5).</t>
            </li>
            <li>
              <t>The closure agent completes the transfer and reports the result.
If no response had arrived before the timeout, it would have
halted the transfer and treated the request as unresolved (B2-4).</t>
            </li>
          </ol>
        </section>
        <section anchor="b2-protocol-requirements">
          <name>Additional Protocol Requirements</name>
          <t>This use case builds on the requirements defined for <xref target="orchestrator-agent"/>
and introduces additional requirements specific to authorization checkpoints in delegated tasks.</t>
          <table>
            <thead>
              <tr>
                <th align="left">REQ-ID</th>
                <th align="left">Description</th>
                <th align="left">Tag</th>
              </tr>
            </thead>
            <tbody>
              <tr>
                <td align="left">B2-1</td>
                <td align="left">The protocol is required to support agent-initiated progress notifications to the delegating agent during task execution.</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">B2-2</td>
                <td align="left">The protocol is required to define an authorization checkpoint request, by which an agent pauses task execution and asks the orchestrator to authorize a specific action before proceeding. The request is required to describe the action to be performed and the effect it would have, so the orchestrator has enough information to decide.</td>
                <td align="left">Transport, Security</td>
              </tr>
              <tr>
                <td align="left">B2-3</td>
                <td align="left">The protocol is required to define the responses to an authorization checkpoint request, which include at minimum approve, deny, and approve with modified parameters. A denial is required to be conveyed as an explicit error response.</td>
                <td align="left">Transport, Security</td>
              </tr>
              <tr>
                <td align="left">B2-4</td>
                <td align="left">The protocol is required to support a response timeout, after which the agent treats the request as unresolved and halts the affected subtask.</td>
                <td align="left">Transport</td>
              </tr>
              <tr>
                <td align="left">B2-5</td>
                <td align="left">An authorization checkpoint response is required to authorize only the action in the request it answers, including any approved modification to that action, and not any different or later action.</td>
                <td align="left">Transport, Security</td>
              </tr>
              <tr>
                <td align="left">B2-6</td>
                <td align="left">The protocol is required to support authorization requested and granted at the granularity of a specific operation, so that a sensitive action is authorized individually rather than covered by a broad, pre-existing authorization.</td>
                <td align="left">Transport, Security</td>
              </tr>
            </tbody>
          </table>
        </section>
      </section>
      <section anchor="peer-collaborative">
        <name>Peer Collaborative Multi-Agent Problem Solving</name>
        <section anchor="description-3">
          <name>Description</name>
          <t>A task requires coordinated problem solving across multiple agents,
where no single agent has full authority or capability to complete
the task alone. The agent that receives the initial request
dynamically delegates subtasks to peer agents based on their
advertised capabilities. Any agent may itself delegate further to
other agents and use other tools, forming a dynamic collaboration
graph. Each agent remains opaque to others, collaborating only
through the protocol interface.</t>
          <t>This use case introduces multi-hop delegation chains that are not
present in <xref target="orchestrator-agent"/>. Each agent in the chain may
delegate further to other agents, and authorization scope is required
to be progressively constrained at each hop.</t>
          <t>This use case is described in <xref target="A2A"/>.</t>
        </section>
        <section anchor="interaction-flow-3">
          <name>Interaction Flow</name>
          <artwork><![CDATA[
  +------------------+
  | Initiating Agent |
  +------------------+
      |         |
      v         v
 +--------+  +--------+
 |Agent-2 |  |Agent-3 |
 +--------+  +--------+
      |          |
      v          v
 +--------+  +--------+
 |Agent-4 |  | Tools  |
 +--------+  +--------+
]]></artwork>
        </section>
        <section anchor="example-3">
          <name>Example</name>
          <t>A hiring manager asks a company's hiring agent to find candidates for
a job listing. The hiring agent delegates sourcing to a recruitment
agency's sourcing agent. The sourcing agent searches a job board tool
and delegates screening of the results to a screening agent, which
assesses each candidate against the listing. The hiring agent later
delegates interviews to a scheduling agent. Each agent is run by a
different provider, in its own administrative domain.</t>
          <artwork><![CDATA[
+--------+    +----------+    +-----------+
| Hiring |--->| Sourcing |--->| Screening |
| agent  |    |  agent   |    |   agent   |
+--------+    +----------+    +-----------+
    |              |
    v              v
+------------+  +-----------+
| Scheduling |  | Job board |
|   agent    |  |   tool    |
+------------+  +-----------+
]]></artwork>
          <ol spacing="normal" type="1"><li>
              <t>The hiring agent establishes a dialog with the sourcing agent and
delegates the search, with an authorization limited to this job
listing (B3-1).</t>
            </li>
            <li>
              <t>The sourcing agent searches the job board tool, then delegates
screening to the screening agent. Because the screening agent is in
another domain, the sourcing agent creates a new dialog linked to
the one it received (B3-9). It passes an authorization narrowed to
reading the candidate profiles for this listing (B3-1, B3-8). The
screening agent can determine that the hiring manager originally
authorized the task (B3-2).</t>
            </li>
            <li>
              <t>If the screening agent becomes unavailable, the sourcing agent
detects this and halts that subtask (B3-5).</t>
            </li>
            <li>
              <t>The hiring manager picks candidates, and the hiring agent delegates
interview scheduling to the scheduling agent, in a new dialog
linked to the first.</t>
            </li>
          </ol>
        </section>
        <section anchor="b3-protocol-requirements">
          <name>Additional Protocol Requirements</name>
          <t>This use case builds on the requirements defined for <xref target="orchestrator-agent"/>
and introduces additional requirements specific to multi-hop delegation chains.</t>
          <table>
            <thead>
              <tr>
                <th align="left">REQ-ID</th>
                <th align="left">Description</th>
                <th align="left">Tag</th>
              </tr>
            </thead>
            <tbody>
              <tr>
                <td align="left">B3-1</td>
                <td align="left">The protocol is required to support multi-hop delegation chains, where an agent that receives a delegated subtask may itself delegate further to other agents. At each hop, the delegating agent is required to present a credential that does not exceed the authorization scope of the credential it received.</td>
                <td align="left">Authentication, Security</td>
              </tr>
              <tr>
                <td align="left">B3-2</td>
                <td align="left">The protocol is required to preserve the identity of the originating entity across all hops in the delegation chain, such that any agent in the chain can determine the identity of the entity that originally authorized the task.</td>
                <td align="left">Authentication, Security</td>
              </tr>
              <tr>
                <td align="left">B3-3</td>
                <td align="left">The protocol is required to support transferable credentials that carry the original authorization constraints across all hops in the delegation chain. Each receiving agent is required to be able to cryptographically verify that the credential presented to it was issued by the delegating agent and that the chain of delegation traces back to the original authorization.</td>
                <td align="left">Authentication, Security</td>
              </tr>
              <tr>
                <td align="left">B3-4</td>
                <td align="left">The protocol is required to ensure that authorization granted to an agent in a delegation chain, including for tool invocations, is derived from the authorization issued by the initiating agent, and not from the identity or authorization scope of any intermediate agent in the chain.</td>
                <td align="left">Authentication, Security</td>
              </tr>
              <tr>
                <td align="left">B3-5</td>
                <td align="left">The protocol is required to allow a delegating agent to detect that a delegated agent has become unavailable and to halt the affected in-flight subtask.</td>
                <td align="left">Dialog, Transport</td>
              </tr>
              <tr>
                <td align="left">B3-6</td>
                <td align="left">The protocol is required to provide a verifiable provenance record of a request and its response as they traverse the delegation chain, attributing each transformation to the hop that performed it, such that removal of a hop, or a modification not attributable to a signing hop, is detectable.</td>
                <td align="left">Authentication, Security</td>
              </tr>
              <tr>
                <td align="left">B3-7</td>
                <td align="left">The protocol is required to enable a receiving party to detect and reject a delegation chain that contains a cycle, such as a chain in which the receiving party's own identifier already appears.</td>
                <td align="left">Authentication, Security</td>
              </tr>
              <tr>
                <td align="left">B3-8</td>
                <td align="left">The protocol is required to enable a receiving party to determine that the authority conveyed to it does not exceed the authority conveyed to the hop it received from.</td>
                <td align="left">Authentication, Security</td>
              </tr>
              <tr>
                <td align="left">B3-9</td>
                <td align="left">The protocol is required to propagate the dialog context across each hop of a delegation chain, so that participants can correlate interactions along the chain. A participant may instead create a new dialog, with a new dialog context, linked to the one it received.</td>
                <td align="left">Dialog</td>
              </tr>
              <tr>
                <td align="left">B3-10</td>
                <td align="left">The protocol is required to propagate cancellation of a delegated subtask, and termination of a dialog, along the delegation chain, and to report to the party that initiated it whether the cancellation or termination took effect further down the chain, or that this could not be determined.</td>
                <td align="left">Dialog</td>
              </tr>
            </tbody>
          </table>
        </section>
      </section>
      <section anchor="cooperative-reasoning">
        <name>Cooperative Reasoning and Consensus Formation</name>
        <section anchor="description-4">
          <name>Description</name>
          <t>A set of peer agents is tasked with analyzing a shared problem
independently and exchanging intermediate reasoning outputs to
converge on a collective conclusion. A coordinator agent distributes
the problem to all participating agents, collects their reasoning
outputs, and drives the convergence process across multiple rounds
until a consensus conclusion is reached. Unlike
<xref target="peer-collaborative"/>, all agents work on the same problem rather
than different subtasks.</t>
          <t>Two communication topologies are possible. In the first, agents
communicate only through the coordinator, which acts as the central
hub for all message exchange. In the second, agents may also
communicate directly with each other to exchange intermediate
reasoning outputs without routing through the coordinator agent. The
second topology introduces the same multi-hop authorization
requirements defined in <xref target="peer-collaborative"/>.</t>
        </section>
        <section anchor="interaction-flow-4">
          <name>Interaction Flow</name>
          <t>The coordinator-mediated topology:</t>
          <artwork><![CDATA[
                +--------------------+
                | Coordinator agent  |
                +--------------------+
                  /      |      \
                 v       v       v
           +--------+ +--------+ +--------+
           |Agent-1 | |Agent-2 | |Agent-3 |
           +--------+ +--------+ +--------+
]]></artwork>
          <t>The direct agent-to-agent topology:</t>
          <artwork><![CDATA[
                 +--------------------+
                 | Coordinator agent  |
                 +--------------------+
                 /          |         \
                v           v          v
           +-------+     +-------+     +-------+
           |Agent-1| <-> |Agent-2| <-> |Agent-3|
           +-------+     +-------+     +-------+
               ^                           ^
               |___________________________|
]]></artwork>
        </section>
        <section anchor="example-4">
          <name>Example</name>
          <t>A research team uses a multi-agent research assistant <xref target="COSCIENTIST"/> to
propose hypotheses for a research goal. A supervisor agent, acting as
the coordinator, distributes the goal to generation agents that
propose hypotheses, a reflection agent that reviews them, and a
ranking agent that ranks them.</t>
          <artwork><![CDATA[
                 +------------------+
                 | Supervisor agent |
                 +------------------+
                  /       |        \
                 v        v         v
        +------------+ +------------+ +---------+
        | Generation | | Reflection | | Ranking |
        |   agents   | |   agent    | |  agent  |
        +------------+ +------------+ +---------+
]]></artwork>
          <ol spacing="normal" type="1"><li>
              <t>The supervisor establishes a separate dialog with each agent and
sends it the research goal.</t>
            </li>
            <li>
              <t>The generation agents propose hypotheses independently.</t>
            </li>
            <li>
              <t>The reflection agent reviews each hypothesis, and the ranking agent
ranks them. The supervisor shares the results with all agents for
the next round.</t>
            </li>
            <li>
              <t>After several rounds, the supervisor returns the top-ranked
hypotheses to the research team and terminates the dialog
(CMN-11).</t>
            </li>
          </ol>
        </section>
      </section>
      <section anchor="tool-mediation">
        <name>Tool, Data, and API Mediation Between Agents</name>
        <section anchor="description-5">
          <name>Description</name>
          <t>In many multi-agent deployments, access to external resources such as
APIs, databases, enterprise systems, or hardware interfaces is
intentionally mediated through a designated mediator. Other agents
request the mediator to perform actions or retrieve data on their
behalf, rather than directly invoking external systems. This
architecture allows access control, auditing, rate limiting, and
schema normalization to be applied uniformly at the mediation layer.</t>
          <t>The mediator may also serve as an adapter between the agent
protocol and non-agent systems or other services that do not natively
support agent communication protocols, or between different agent
communication protocols such as translating between the agentic
protocol suite being developed at the IETF and existing protocols
such as <xref target="MCP"/> and <xref target="A2A"/>. In this role, the mediator is responsible
for protocol translation and for presenting the appropriate
credentials to the target system on behalf of the requesting agent.</t>
          <t>The mediator may additionally act as a request router, dispatching requests
to appropriate agents or tools based on the content and context of
the request, without requiring the requesting agent to have prior
knowledge of which agent or tool is most appropriate.</t>
          <t>The mediator may also validate agent requests before invocation,
checking whether the action being requested matches the
authorization granted to the agent and whether execution would
cause unintended or irreversible side effects.</t>
          <t>Note that mediating can be a function within an orchestrator.</t>
          <t>The mediator can be deployed as a proxy between an agent and its
tools, such as an MCP proxy. The 3GPP 6G architecture study
<xref target="TR23801-01"/> describes it as an agent gateway through which
third-party agents use network agents.</t>
          <t>This pattern is reflected in the MCP server architecture defined
in <xref target="MCP"/> and the agent routing patterns discussed in <xref target="A2A"/>.</t>
        </section>
        <section anchor="interaction-flow-5">
          <name>Interaction Flow</name>
          <artwork><![CDATA[
       +--------------+
       |    Agent     |
       +--------------+
              |
              v
       +--------------+
       |   Mediator   |
       +--------------+
       /       |       \
      v        v        v
+--------+ +---------+  +--------+
| Agent-1 | | Tool   |  | Agent-2 |
+--------+ +---------+  +--------+
]]></artwork>
        </section>
        <section anchor="example-5">
          <name>Example</name>
          <t>A mobile network operator runs an agent gateway through which
authorized third-party agents use network agents. A concert organizer's agent asks for extra network
capacity at the venue during the concert.</t>
          <artwork><![CDATA[
+-------------+      +----------+      +--------------------+
| Organizer's |----->|  Agent   |----->| Network agent      |
| agent       |      | gateway  |      |                    |
+-------------+      +----------+      +--------------------+
]]></artwork>
          <ol spacing="normal" type="1"><li>
              <t>The organizer's agent establishes a dialog with the gateway and
requests extra capacity at the venue for the hours of the concert.</t>
            </li>
            <li>
              <t>The gateway checks that the request matches the organizer's
authorization and forwards it to the network agent.</t>
            </li>
            <li>
              <t>The gateway confirms the added capacity to the organizer's agent.</t>
            </li>
            <li>
              <t>After the concert, the organizer's agent terminates the dialog
(CMN-11).</t>
            </li>
          </ol>
        </section>
        <section anchor="b5-protocol-requirements">
          <name>Additional Protocol Requirements</name>
          <table>
            <thead>
              <tr>
                <th align="left">REQ-ID</th>
                <th align="left">Description</th>
                <th align="left">Tag</th>
              </tr>
            </thead>
            <tbody>
              <tr>
                <td align="left">B5-1</td>
                <td align="left">The protocol is required to define error response types for request validation failure (rejected due to potential unintended or irreversible side effects) and protocol translation failure (rejected on unsuccessful translation of a request or response between supported protocols), distinct from authorization failure.</td>
                <td align="left">Transport, Security</td>
              </tr>
              <tr>
                <td align="left">B5-2</td>
                <td align="left">The protocol is required to support continuation of a dialog through a different mediator when the mediator in use fails.</td>
                <td align="left">Dialog</td>
              </tr>
              <tr>
                <td align="left">B5-3</td>
                <td align="left">The protocol is required to enable a mediator that translates between agent communication protocols to carry the dialog context across the translation, so that participants on each side can correlate the dialog.</td>
                <td align="left">Dialog</td>
              </tr>
            </tbody>
          </table>
        </section>
      </section>
    </section>
    <section anchor="relationship-oauth-wimse">
      <name>Relationship to OAuth and WIMSE Work</name>
      <t>Several requirements in this document concern authorization, delegation, and agent identity (CMN-4, CMN-8, CMN-9, CMN-10, B2-2 to B2-6, B3-1 to B3-4, B3-6, B3-7, B3-8). These are expected to be addressed in the OAuth and WIMSE Working Groups.</t>
      <t>Rejecting a dialog change or a resumption when a check fails or cannot be completed (CMN-12, B1-7, B1-8) is a requirement for the dialog management protocol. The check itself, verifying identity and authorization, is expected to be addressed in the OAuth and WIMSE Working Groups.</t>
      <t>Authorizing an individual action within a task, including when the authorization information is unavailable or may no longer be current, is also expected to be addressed in the OAuth Working Group.</t>
    </section>
    <section anchor="gaps">
      <name>Existing Protocols and Gaps</name>
      <t><xref target="A2A"/> groups related tasks under a context identifier, and <xref target="MCP"/>
groups the requests between one client and one server under a session.
Both are scoped to a single hop. An agent that delegates to another
agent or calls a tool starts a new context or session, and no
identifier spans the hops. Cancellation is per task and per hop.</t>
      <t>The gap is end-to-end: correlating a task across hops, protocols, and
connections, and propagating lifecycle actions such as cancellation
along the delegation chain. These are the dialog requirements in this
document (CMN-11, CMN-12, B1-6 to B1-8, B3-9, B3-10, B5-2, B5-3). They
can be met by carrying the dialog context across hops or by linked
dialogs.</t>
      <t>This analysis reflects <xref target="A2A"/> and <xref target="MCP"/> at the time of writing.</t>
    </section>
    <section anchor="accountability-auditing">
      <name>Accountability and Auditing</name>
      <t>Accountability and auditing of agent actions are out of scope for this
document. Audit and provenance representation is addressed by existing work,
including W3C provenance (PROV) and Trace Context. Dialog identifiers can be
used to correlate audit records across the participants in a dialog.</t>
      <t>Editor's note: Add a reference to IETF audit work once it progresses.</t>
    </section>
    <section anchor="security">
      <name>Security Considerations</name>
      <t>Security considerations are addressed throughout this document via
the Identity, Authentication, and Delegation requirements defined
for each use case.</t>
      <t>Agent identity information is considered to be sensitive, particularly
in multi-domain deployments. Use of persistent identifiers across
dialogs and domains can enable tracking and correlation of agent
activity. Protocol designers need to consider mechanisms such as
pseudonymous or temporary identifiers to reduce linkability,
while preserving the ability to audit and enforce accountability
where required. The trade-offs between privacy, accountability,
and traceability need to be considered in the design of agent identity mechanisms.</t>
      <t>Agent unavailability (B3-5) halts the affected subtask but does not imply
revocation, since it may be a transient failure rather than compromise.</t>
      <t>Signaled error reasons help a participant decide on the right course
of action to address the error. They need to be precise enough for
that, but not reveal information that an attacker could use.</t>
    </section>
    <section anchor="iana">
      <name>IANA Considerations</name>
      <t>This document has no IANA actions.</t>
    </section>
    <section numbered="false" anchor="acknowledgements">
      <name>Acknowledgements</name>
      <t>Thanks to Borislava Gajic, Julien Maisonneuve, Parisa Foroughi, Laurent Ciavaglia, Nathalie Romo Moreno, Peter Leis, Linda Dunbar, Mikhail Sergeev, Iman Schrock, Sumit Ahuja, Sina Khatibi, Vaishnavi Gudur, Lars Kersten Kroehl, Girish Konda, Nikhil Donapati, Nicholas Templeman, Bradley B, Douglas Wadkins, and Lionel Morand for the discussions and comments.</t>
      <t>The examples in <xref target="simple-single-agent"/> and <xref target="authz-checkpoint"/> are
based on scenarios in draft-rosenberg-agentproto-usecases by Jonathan
Rosenberg and Cullen Jennings.</t>
    </section>
  </middle>
  <back>
    <references anchor="sec-informative-references">
      <name>Informative References</name>
      <reference anchor="A2A" target="https://a2a-protocol.org/latest/specification/">
        <front>
          <title>Agent2Agent Protocol Specification</title>
          <author>
            <organization/>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
      <reference anchor="MCP" target="https://modelcontextprotocol.io/specification/2025-11-25">
        <front>
          <title>Model Context Protocol Specification</title>
          <author>
            <organization/>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
      <reference anchor="SCRM" target="https://datatracker.ietf.org/doc/draft-scrm-aiproto-usecases">
        <front>
          <title>Agentic AI Use Cases</title>
          <author>
            <organization/>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
      <reference anchor="COSCIENTIST" target="https://arxiv.org/abs/2502.18864">
        <front>
          <title>Accelerating scientific discovery with Co-Scientist</title>
          <author>
            <organization>Google DeepMind</organization>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
      <reference anchor="TR23801-01" target="https://www.3gpp.org/ftp/Specs/archive/23_series/23.801-01/">
        <front>
          <title>Study on Architecture for 6G System</title>
          <author>
            <organization>3GPP</organization>
          </author>
          <date>n.d.</date>
        </front>
        <seriesInfo name="3GPP" value="TR 23.801-01 v0.9.0"/>
      </reference>
      <reference anchor="YAO" target="https://datatracker.ietf.org/doc/draft-yao-catalist-problem-space-analysis">
        <front>
          <title>Problem Space Analysis of AI Agent Protocols in IETF</title>
          <author>
            <organization/>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
      <reference anchor="SONG" target="https://datatracker.ietf.org/doc/draft-song-dmsc-problem-statement">
        <front>
          <title>Problem Statement and Requirements for Dynamic Multi-agent Secured Collaboration</title>
          <author>
            <organization/>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
      <reference anchor="RFC7696" target="https://www.rfc-editor.org/rfc/rfc7696">
        <front>
          <title>Guidelines for Cryptographic Algorithm Agility and Selecting Mandatory-to-Implement Algorithms</title>
          <author>
            <organization/>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
      <reference anchor="AUTOGEN" target="https://microsoft.github.io/autogen/stable/">
        <front>
          <title>AutoGen: A Framework for Multi-Agent Conversation</title>
          <author>
            <organization/>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
      <reference anchor="LANGCHAIN" target="https://python.langchain.com/docs/concepts/agents/">
        <front>
          <title>LangChain Agent Framework</title>
          <author>
            <organization/>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
      <reference anchor="OPENAI-AGENTS" target="https://openai.github.io/openai-agents-python/">
        <front>
          <title>OpenAI Agents SDK</title>
          <author>
            <organization/>
          </author>
          <date>n.d.</date>
        </front>
      </reference>
    </references>
  </back>
  <!-- ##markdown-source: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-->

</rfc>
