haive.games.mafia.aug_llms
==========================

.. py:module:: haive.games.mafia.aug_llms

.. autoapi-nested-parse::

   Specialized LLM configurations for the Mafia game.

   This module provides specialized augmented LLM configurations for different
   aspects of the Mafia game, including:
       - Role-specific analyzer LLMs
       - Strategic decision-making models
       - Game state evaluators

   These configurations extend the basic engines.py configurations with more
   sophisticated models tailored for specific game aspects.

   .. rubric:: Example

   >>> from haive.games.mafia.aug_llms import get_mafia_analyzer
   >>>
   >>> # Get an analyzer for evaluating player suspicion levels
   >>> analyzer = get_mafia_analyzer("suspicion")
   >>> analysis = analyzer.invoke(game_state)



Functions
---------

.. autoapisummary::

   haive.games.mafia.aug_llms.get_mafia_analyzer


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

.. py:function:: get_mafia_analyzer(analyzer_type)

   Get a specialized Mafia game analyzer.

   This function returns a configured analyzer LLM for specific
   Mafia game analysis tasks, such as suspicion evaluation,
   player psychology, strategy optimization, or voting analysis.

   :param analyzer_type: Type of analyzer to get ("suspicion", "psychology",
                         "strategy", or "voting")

   :returns: Configured analyzer
   :rtype: AugLLMConfig

   :raises ValueError: If analyzer_type is not recognized

   .. rubric:: Example

   >>> analyzer = get_mafia_analyzer("suspicion")
   >>> analysis = analyzer.invoke(game_context)