haive.agents.supervisor.utils.routing
=====================================

.. py:module:: haive.agents.supervisor.utils.routing

.. autoapi-nested-parse::

   Dynamic Routing Engine for Haive Supervisor System.

   Handles intelligent routing decisions using DynamicChoiceModel and LLM-based analysis.
   Provides context-aware agent selection with validation and fallback mechanisms.



Classes
-------

.. autoapisummary::

   haive.agents.supervisor.utils.routing.BaseRoutingStrategy
   haive.agents.supervisor.utils.routing.DynamicRoutingEngine
   haive.agents.supervisor.utils.routing.LLMRoutingStrategy
   haive.agents.supervisor.utils.routing.RoutingContext
   haive.agents.supervisor.utils.routing.RoutingDecision
   haive.agents.supervisor.utils.routing.RuleBasedRoutingStrategy
   haive.agents.supervisor.utils.routing.TaskClassifier


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

.. py:class:: BaseRoutingStrategy

   Bases: :py:obj:`abc.ABC`


   Abstract base for routing strategies.


   .. py:method:: make_routing_decision(context, available_agents, agent_capabilities, config = None)
      :abstractmethod:

      :async:


      Make routing decision based on context.



.. py:class:: DynamicRoutingEngine(routing_model, routing_engine = None, routing_strategy = None, enable_context_analysis = True)

   Main routing engine that orchestrates routing decisions.

   Handles context extraction, strategy selection, and routing execution
   with comprehensive error handling and fallback mechanisms.

   Initialize routing engine.

   :param routing_model: DynamicChoiceModel for available choices
   :param routing_engine: Engine for LLM-based routing (optional)
   :param routing_strategy: Custom routing strategy (optional)
   :param enable_context_analysis: Whether to perform context analysis


   .. py:method:: print_routing_stats()

      Print routing engine statistics.



   .. py:method:: route_request(state, agent_capabilities, config = None)
      :async:


      Main routing method.

      :param state: Current graph state
      :param agent_capabilities: Dict of agent capabilities
      :param config: Runnable configuration

      :returns: Command object with routing decision



.. py:class:: LLMRoutingStrategy(routing_engine, routing_model)

   Bases: :py:obj:`BaseRoutingStrategy`


   LLM-based routing strategy using structured output.

   Init  .

   :param routing_engine: [TODO: Add description]
   :param routing_model: [TODO: Add description]


   .. py:method:: make_routing_decision(context, available_agents, agent_capabilities, config = None)
      :async:


      Make LLM-based routing decision.



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

   Bases: :py:obj:`pydantic.BaseModel`


   Context extracted from state for routing decisions.

   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:class:: RoutingDecision(/, **data)

   Bases: :py:obj:`pydantic.BaseModel`


   Model for routing decisions.

   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:class:: RuleBasedRoutingStrategy(routing_rules)

   Bases: :py:obj:`BaseRoutingStrategy`


   Rule-based routing strategy for deterministic routing.

   Initialize with routing rules.

   :param routing_rules: Dict mapping keywords to agent names


   .. py:method:: make_routing_decision(context, available_agents, agent_capabilities, config = None)
      :async:


      Make rule-based routing decision.



.. py:class:: TaskClassifier

   Classifies tasks for better routing decisions.


   .. py:method:: classify_task(message)
      :classmethod:


      Classify task based on message content.

      :param message: Message to classify

      :returns: List of detected task types



   .. py:method:: estimate_complexity(message, conversation_length)
      :classmethod:


      Estimate task complexity.

      :param message: Message content
      :param conversation_length: Length of conversation

      :returns: Simple/Medium/Complex
      :rtype: Complexity level