agents.rag.simple.enhanced_v3.state

Enhanced RAG State Schema for SimpleRAG V3.

This module provides enhanced state management for SimpleRAG using Enhanced MultiAgent V3 with performance tracking, debug information, and comprehensive metadata.

Classes

GenerationDebugInfo

Debug information for generation operations.

RAGMetadata

Metadata for RAG operations.

RetrievalDebugInfo

Debug information for retrieval operations.

SimpleRAGState

Enhanced state schema for SimpleRAG V3 pipeline.

Module Contents

class agents.rag.simple.enhanced_v3.state.GenerationDebugInfo(/, **data)

Bases: pydantic.BaseModel

Debug information for generation operations.

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.

Parameters:

data (Any)

class agents.rag.simple.enhanced_v3.state.RAGMetadata(/, **data)

Bases: pydantic.BaseModel

Metadata for RAG operations.

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.

Parameters:

data (Any)

class agents.rag.simple.enhanced_v3.state.RetrievalDebugInfo(/, **data)

Bases: pydantic.BaseModel

Debug information for retrieval operations.

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.

Parameters:

data (Any)

class agents.rag.simple.enhanced_v3.state.SimpleRAGState(/, **data)

Bases: haive.core.schema.state_schema.StateSchema

Enhanced state schema for SimpleRAG V3 pipeline.

This state schema extends the basic StateSchema with RAG-specific fields and enhanced tracking capabilities when performance_mode or debug_mode are enabled.

Core RAG Fields:

  • query: User query string

  • retrieved_documents: Documents from retrieval step

  • generated_answer: Final answer from generation step

Enhanced Tracking (when enabled):

  • retrieval_metadata: Retrieval operation metadata

  • generation_metadata: Generation operation metadata

  • performance_metrics: Performance tracking data

  • debug_info: Detailed debug information

Examples

Basic usage (automatic schema selection):

# Enhanced features disabled - uses basic fields only
state = SimpleRAGState(query="What is AI?")

Enhanced usage:

# Enhanced features enabled - includes all tracking
state = SimpleRAGState(
    query="What is AI?",
    retrieval_metadata=RAGMetadata(
        timing_info={"retrieval_time": 0.5}
    ),
    debug_mode=True
)

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.

Parameters:

data (Any)

add_generation_debug(context_length=None, prompt_tokens=None, completion_tokens=None, generation_time=None, **kwargs)

Add generation debug information.

Parameters:

  • context_length (int | None)

  • prompt_tokens (int | None)

  • completion_tokens (int | None)

  • generation_time (float | None)

Return type:

None

add_retrieval_debug(query_vector_dim=None, search_time=None, total_documents=None, similarity_scores=None, **kwargs)

Add retrieval debug information.

Parameters:

  • query_vector_dim (int | None)

  • search_time (float | None)

  • total_documents (int | None)

  • similarity_scores (list[float] | None)

Return type:

None

get_generation_summary()

Get generation operation summary.

Return type:

dict[str, Any]

get_pipeline_summary()

Get comprehensive pipeline summary.

Return type:

dict[str, Any]

get_retrieval_summary()

Get retrieval operation summary.

Return type:

dict[str, Any]

update_performance_metric(metric_name, value)

Update a performance metric.

Parameters:
Return type:

None

update_stage(stage)

Update current stage and add to history.

Parameters:

stage (str)

Return type:

None