haive.agents.graphs
===================

.. py:module:: haive.agents.graphs

.. autoapi-nested-parse::

   LangGraph Platform graph entry points for haive-agents.

   This module exposes all workable agent graphs as callable factories for
   LangGraph Platform (langgraph.json). Each function returns a compiled
   LangGraph StateGraph served via ``langgraph dev`` or LangGraph Cloud.



Functions
---------

.. autoapisummary::

   haive.agents.graphs.collaborative_conversation
   haive.agents.graphs.debate_conversation
   haive.agents.graphs.document_grading_rag
   haive.agents.graphs.dynamic_rag
   haive.agents.graphs.flare_rag
   haive.agents.graphs.fusion_rag
   haive.agents.graphs.lats
   haive.agents.graphs.llm_compiler
   haive.agents.graphs.llm_rag
   haive.agents.graphs.memory_agent
   haive.agents.graphs.multi_agent_sequential
   haive.agents.graphs.multi_query_rag
   haive.agents.graphs.plan_and_execute
   haive.agents.graphs.rag_agent
   haive.agents.graphs.react_agent
   haive.agents.graphs.reasoning_system
   haive.agents.graphs.reflection_agent
   haive.agents.graphs.reflexion
   haive.agents.graphs.rewoo
   haive.agents.graphs.round_robin_conversation
   haive.agents.graphs.self_discover_adapter
   haive.agents.graphs.self_discover_executor
   haive.agents.graphs.self_discover_selector
   haive.agents.graphs.self_discover_structurer
   haive.agents.graphs.self_route_rag
   haive.agents.graphs.simple_agent
   haive.agents.graphs.simple_rag
   haive.agents.graphs.speculative_rag
   haive.agents.graphs.step_back_rag
   haive.agents.graphs.supervisor
   haive.agents.graphs.task_analysis


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

.. py:function:: collaborative_conversation()

   Collaborative multi-agent conversation.


.. py:function:: debate_conversation()

   Structured debate between agents.


.. py:function:: document_grading_rag()

   Document Grading RAG: grades retrieved documents for relevance.


.. py:function:: dynamic_rag()

   Dynamic RAG with multiple retrieval sources.


.. py:function:: flare_rag()

   FLARE RAG: forward-looking active retrieval.


.. py:function:: fusion_rag()

   RAG Fusion: reciprocal rank fusion across multiple retrievals.


.. py:function:: lats()

   LATS: Monte Carlo Tree Search over LLM responses with reflection scoring.


.. py:function:: llm_compiler()

   LLM Compiler: structured DAG planner + parallel executor + joiner.


.. py:function:: llm_rag()

   LLM-based RAG agent.


.. py:function:: memory_agent()

   Memory-enhanced agent with vector store + knowledge graph.


.. py:function:: multi_agent_sequential()

   Sequential multi-agent pipeline (agent A -> B -> C).


.. py:function:: multi_query_rag()

   Multi-Query RAG: generates multiple query variants for retrieval.


.. py:function:: plan_and_execute()

   Plan-and-execute agent: creates a plan then executes each step.


.. py:function:: rag_agent()

   Base retrieval-augmented generation agent (HuggingFace embeddings).


.. py:function:: react_agent()

   ReAct agent with reasoning + acting loop and tool-use.


.. py:function:: reasoning_system()

   Logic-based reasoning system with multiple engines.


.. py:function:: reflection_agent()

   Self-reflection agent that generates then improves its answer.


.. py:function:: reflexion()

   Reflexion: draft -> reflect -> revise loop with ReactAgent + SimpleAgent.


.. py:function:: rewoo()

   ReWOO: plan all steps upfront, execute in parallel, synthesize answer.


.. py:function:: round_robin_conversation()

   Round-robin conversation: each agent speaks in fixed order.


.. py:function:: self_discover_adapter()

   Self-Discover adapter: adapts modules to the task.


.. py:function:: self_discover_executor()

   Self-Discover executor: executes the structured reasoning plan.


.. py:function:: self_discover_selector()

   Self-Discover selector: picks relevant reasoning modules.


.. py:function:: self_discover_structurer()

   Self-Discover structurer: structures the reasoning plan.


.. py:function:: self_route_rag()

   Self-Route RAG: analyzes query to choose optimal retrieval strategy.


.. py:function:: simple_agent()

   Basic conversational agent with hooks and structured output.


.. py:function:: simple_rag()

   Simple RAG = retriever + answer agent in sequence.


.. py:function:: speculative_rag()

   Speculative RAG: hypothesis generation + parallel verification.


.. py:function:: step_back_rag()

   Step-Back RAG: generates abstracted queries for better retrieval.


.. py:function:: supervisor()

   Dynamic supervisor that routes tasks to sub-agents.


.. py:function:: task_analysis()

   Task analysis and decomposition agent.