agents.rag.multi_agent_rag.graded_rag_workflows_v2¶

Graded RAG Workflows V2 - Using Enhanced State Schemas.

This version uses state schemas with built-in configuration support, providing a cleaner approach to managing agent-specific parameters.

Classes¶

`FLAREAgentV2Example`	FLARE Agent V2 example using enhanced state schema.
`FullyGradedRAGAgentV2`	Fully Graded RAG V2 - Uses enhanced state schema with configuration support.
`MultiCriteriaGradedRAGAgentV2`	Multi-Criteria Graded RAG V2 - Configuration stored in state schema.

Functions¶

`build_custom_graph`()	Build custom graph for graded RAG workflows v2.
`create_graded_rag_agent`([workflow_type, ...])	Factory function to create graded RAG agents with proper configuration.

Module Contents¶

class agents.rag.multi_agent_rag.graded_rag_workflows_v2.FLAREAgentV2Example(uncertainty_threshold=0.3, max_retrieval_rounds=3, **kwargs)¶

Bases: haive.agents.multi.base.MultiAgent, haive.agents.rag.multi_agent_rag.enhanced_state_schemas.StateConfigMixin

FLARE Agent V2 example using enhanced state schema.

Init .

Parameters:

uncertainty_threshold (float) – [TODO: Add description]
max_retrieval_rounds (int) – [TODO: Add description]

build_custom_graph()¶

Build custom graph.

Return type:: Any

class agents.rag.multi_agent_rag.graded_rag_workflows_v2.FullyGradedRAGAgentV2(relevance_threshold=0.5, **kwargs)¶

Bases: haive.agents.multi.base.MultiAgent, haive.agents.rag.multi_agent_rag.enhanced_state_schemas.StateConfigMixin

Fully Graded RAG V2 - Uses enhanced state schema with configuration support.

Init .

Parameters:: relevance_threshold (float) – [TODO: Add description]

async ainvoke(inputs)¶

Override to inject configuration into state.

Parameters:: inputs (dict[str, Any])
Return type:: dict[str, Any]

build_custom_graph()¶

Build the custom graph with state initialization.

Return type:: Any

class agents.rag.multi_agent_rag.graded_rag_workflows_v2.MultiCriteriaGradedRAGAgentV2(grading_criteria=None, **kwargs)¶

Bases: haive.agents.multi.base.MultiAgent, haive.agents.rag.multi_agent_rag.enhanced_state_schemas.StateConfigMixin

Multi-Criteria Graded RAG V2 - Configuration stored in state schema.

Init .

Parameters:: grading_criteria (list[str] | None) – [TODO: Add description]

async ainvoke(inputs)¶

Override to inject configuration.

Parameters:: inputs (dict[str, Any])
Return type:: dict[str, Any]

agents.rag.multi_agent_rag.graded_rag_workflows_v2.build_custom_graph()¶

Build custom graph for graded RAG workflows v2.

Returns:: Graph configuration or None for default behavior
Return type:: Any

agents.rag.multi_agent_rag.graded_rag_workflows_v2.create_graded_rag_agent(workflow_type='fully_graded', relevance_threshold=0.5, grading_criteria=None, **kwargs)¶

Factory function to create graded RAG agents with proper configuration.

Parameters:

workflow_type (str)
relevance_threshold (float)
grading_criteria (list[str] | None)

Return type:

haive.agents.multi.base.MultiAgent