haive.agents.dynamic_supervisor.agent
=====================================

.. py:module:: haive.agents.dynamic_supervisor.agent

.. autoapi-nested-parse::

   Dynamic Supervisor Agent - routes tasks to sub-agents via tool calls.

   The supervisor extends ReactAgent. Each sub-agent becomes a handoff tool.
   The LLM decides which agent to call by making tool calls. Adding or removing
   agents triggers graph recompilation so the LLM always sees current tools.

   Example::

       supervisor = DynamicSupervisor(name="boss", engine=AugLLMConfig())
       supervisor.add_agent(math_agent, "Solves math problems")
       supervisor.add_agent(writer_agent, "Writes creative content")
       result = supervisor.run("What is 42 * 17?")  # routes to math_agent



Classes
-------

.. autoapisummary::

   haive.agents.dynamic_supervisor.agent.DynamicSupervisor


Module Contents
---------------

.. py:class:: DynamicSupervisor(/, **data)

   Bases: :py:obj:`haive.agents.react.agent.ReactAgent`


   Supervisor that routes tasks to sub-agents via tool calls.

   Agents are added at runtime with add_agent(). Each becomes a
   handoff_to_{name} tool. The ReactAgent loop handles routing -
   the LLM makes tool calls to delegate work.

   Create a new model by parsing and validating input data from keyword arguments.

   Raises [`ValidationError`][pydantic_core.ValidationError] if the input data cannot be
   validated to form a valid model.

   `self` is explicitly positional-only to allow `self` as a field name.


   .. py:method:: add_agent(agent, description = None, name = None)

      Add a sub-agent. Creates a handoff tool and recompiles the graph.

      :param agent: The agent to add (must have .run() or .arun())
      :param description: What this agent does (shown to LLM)
      :param name: Override agent name (defaults to agent.name)



   .. py:method:: compile(**kwargs)

      Compile using the LangGraph path (not SimpleAgent's create_runnable).

      SimpleAgent.compile() uses create_runnable() which doesn't rebuild
      properly after tool changes. We go through BaseGraph.to_langgraph()
      instead.



   .. py:method:: create_agent(name, system_message, description, tools = None, temperature = 0.7)

      Create a new agent on the fly and add it to the supervisor.

      :param name: Agent name
      :param system_message: System prompt for the new agent
      :param description: What this agent does
      :param tools: Optional tools for the new agent
      :param temperature: LLM temperature

      :returns: The created agent



   .. py:method:: model_post_init(__context)

      Store base system message before super init.



   .. py:method:: remove_agent(name)

      Remove a sub-agent and its handoff tool.



   .. py:property:: agents
      :type: dict[str, Any]


      Get the current agent registry.