agents.multi.experiments.implementations.self_discover_state

State schema for self-discover multi-agent system.

Classes

SelfDiscoverState

State schema for self-discover multi-agent workflow.

Module Contents

class agents.multi.experiments.implementations.self_discover_state.SelfDiscoverState(/, **data)

Bases: haive.core.schema.prebuilt.multi_agent_state.MultiAgentState

State schema for self-discover multi-agent workflow.

This state schema handles the structured output flow between agents: 1. select_agent: reasoning_modules + task_description → selected_modules 2. adapt_agent: selected_modules + task_description → adapted_modules 3. structure_agent: adapted_modules + task_description → reasoning_structure 4. reason_agent: reasoning_structure + task_description → final_answer

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)

get_adapt_inputs()

Get inputs for adapt_agent.

Return type:

dict[str, Any]

get_reason_inputs()

Get inputs for reason_agent.

Return type:

dict[str, Any]

get_select_inputs()

Get inputs for select_agent.

Return type:

dict[str, Any]

get_structure_inputs()

Get inputs for structure_agent.

Return type:

dict[str, Any]

update_from_agent_output(agent_name, output)

Update state with agent output.

Parameters:
Return type:

None