agents.memory_v2.advanced_rag_memory_agent

Advanced RAG Memory Agent with multi-stage retrieval and reranking.

This implementation provides state-of-the-art RAG capabilities: 1. Multi-stage retrieval: dense → sparse → reranking 2. Hybrid search combining vector, keyword, and graph 3. Query decomposition for complex questions 4. Memory-augmented generation with citations 5. Adaptive retrieval based on query complexity

Classes

AdvancedRAGConfig	Configuration for Advanced RAG Memory Agent.
AdvancedRAGMemoryAgent	Advanced RAG Memory Agent with multi-stage retrieval.
QueryComplexity	Query complexity levels.
RetrievalStrategy	Different retrieval strategies available.

Functions

create_conversational_memory_agent()	Create a conversation-focused memory agent.
create_research_memory_agent()	Create a research-focused memory agent.
example_advanced_rag_usage()	Example of using Advanced RAG Memory Agent.

Module Contents

class agents.memory_v2.advanced_rag_memory_agent.AdvancedRAGConfig

Configuration for Advanced RAG Memory Agent.

class agents.memory_v2.advanced_rag_memory_agent.AdvancedRAGMemoryAgent(config)

Advanced RAG Memory Agent with multi-stage retrieval.

This agent implements state-of-the-art retrieval-augmented generation with sophisticated memory management capabilities.

Init .

Parameters:

config (AdvancedRAGConfig) – [TODO: Add description]

async add_memory(content, metadata=None, importance='normal')

Add new memory to the system.

Parameters:

• content (str)

• metadata (dict[str, Any] | None)

• importance (str)

Return type:

dict[str, Any]

analyze_query_complexity(query)

Analyze query complexity to choose optimal strategy.

Parameters:

query (str)

Return type:

QueryComplexity

choose_retrieval_strategy(query, complexity)

Choose optimal retrieval strategy based on query and complexity.

Parameters:

• query (str)

• complexity (QueryComplexity)

Return type:

RetrievalStrategy

async generate_with_citations(query, retrieved_docs, include_citations=None)

Generate response with citations.

Parameters:

• query (str)

• retrieved_docs (list[langchain_core.documents.Document])

• include_citations (bool | None)

Return type:

dict[str, Any]

async get_memory_analytics()

Get comprehensive analytics about memory usage.

Return type:

dict[str, Any]

async query_memory(query, strategy=None, include_analysis=True)

Query memory with advanced RAG capabilities.

Parameters:

• query (str)

• strategy (RetrievalStrategy | None)

• include_analysis (bool)

Return type:

dict[str, Any]

async retrieve_documents(query, strategy=None, k=None)

Retrieve documents using specified strategy.

Parameters:

• query (str)

• strategy (RetrievalStrategy | None)

• k (int | None)

Return type:

list[langchain_core.documents.Document]

save_memory_store(path=None)

Save the vector store and metadata.

Parameters:

path (str | None)

class agents.memory_v2.advanced_rag_memory_agent.QueryComplexity

Bases: str, enum.Enum

Query complexity levels.

Initialize self. See help(type(self)) for accurate signature.

class agents.memory_v2.advanced_rag_memory_agent.RetrievalStrategy

Bases: str, enum.Enum

Different retrieval strategies available.

Initialize self. See help(type(self)) for accurate signature.

async agents.memory_v2.advanced_rag_memory_agent.create_conversational_memory_agent()

Create a conversation-focused memory agent.

Return type:

AdvancedRAGMemoryAgent

async agents.memory_v2.advanced_rag_memory_agent.create_research_memory_agent()

Create a research-focused memory agent.

Return type:

AdvancedRAGMemoryAgent

async agents.memory_v2.advanced_rag_memory_agent.example_advanced_rag_usage()

Example of using Advanced RAG Memory Agent.