self_query_mcp_agent
====================

.. py:module:: self_query_mcp_agent

.. autoapi-nested-parse::

   Self Query MCP Discovery Agent.

   Enhanced RAG agent that uses Self Query methodology for structured querying
   with metadata filtering and parent document retrieval.



Classes
-------

.. autoapisummary::

   self_query_mcp_agent.EnhancedMCPDocument
   self_query_mcp_agent.MCPServerMetadata
   self_query_mcp_agent.SelfQueryMCPAgent


Functions
---------

.. autoapisummary::

   self_query_mcp_agent.test_enhanced_agent


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

.. py:class:: EnhancedMCPDocument

   Enhanced document creation with chunking for parent retrieval.


   .. py:method:: create_mcp_documents_with_chunks() -> tuple[list[langchain.schema.Document], list[langchain.schema.Document]]
      :staticmethod:


      Create both parent documents and child chunks for retrieval.



.. py:class:: MCPServerMetadata(/, **data: Any)

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


   Metadata schema for MCP servers.


   .. py:attribute:: category
      :type:  str
      :value: None



   .. py:attribute:: has_install
      :type:  bool
      :value: None



   .. py:attribute:: language
      :type:  str
      :value: None



   .. py:attribute:: prompts_count
      :type:  int
      :value: None



   .. py:attribute:: resources_count
      :type:  int
      :value: None



   .. py:attribute:: server_name
      :type:  str
      :value: None



   .. py:attribute:: stars
      :type:  int
      :value: None



   .. py:attribute:: tools_count
      :type:  int
      :value: None



   .. py:attribute:: total_features
      :type:  int
      :value: None



.. py:class:: SelfQueryMCPAgent(config: haive.core.engine.aug_llm.AugLLMConfig = None)

   Enhanced MCP Discovery Agent with Self Query and Parent Document Retrieval.


   .. py:method:: analyze_query_intent(query: str) -> str

      Analyze query to determine best search method.



   .. py:method:: hybrid_search(query: str, k: int = 5) -> dict[str, list[langchain.schema.Document]]
      :async:


      Perform all search methods and return results.



   .. py:method:: search_similarity(query: str, k: int = 5) -> list[langchain.schema.Document]
      :async:


      Fallback similarity search.



   .. py:method:: search_with_parent_retriever(query: str, k: int = 5) -> list[langchain.schema.Document]
      :async:


      Search using parent document retriever for full content.



   .. py:method:: search_with_self_query(query: str, k: int = 5) -> list[langchain.schema.Document]
      :async:


      Search using self-query retriever for structured queries.



   .. py:method:: setup_retrievers()

      Set up both self-query and parent document retrievers.



   .. py:attribute:: config


   .. py:attribute:: embeddings


   .. py:attribute:: llm


.. py:function:: test_enhanced_agent()
   :async:


   Test the enhanced MCP agent.