agents.multi.enhanced_sequential_agent¶

Enhanced SequentialAgent implementation using Agent[AugLLMConfig].

SequentialAgent = Agent[AugLLMConfig] + sequential execution of agents.

Classes¶

`MockAgent`	Init .
`SequentialAgent`	Enhanced SequentialAgent that executes agents in sequence.

Module Contents¶

class agents.multi.enhanced_sequential_agent.MockAgent(name, transform)¶

Init .

Parameters:

name (str) – [TODO: Add description]
transform (str) – [TODO: Add description]

async arun(input_data)¶

Arun.

Parameters:: input_data (str) – [TODO: Add description]
Returns:: Add return description]
Return type:: [TODO

class agents.multi.enhanced_sequential_agent.SequentialAgent¶

Bases: haive.agents.simple.enhanced_simple_real.EnhancedAgentBase

Enhanced SequentialAgent that executes agents in sequence.

SequentialAgent = Agent[AugLLMConfig] + sequential pipeline execution.

Each agent’s output becomes the next agent’s input, creating a pipeline. The coordinator (this agent) can optionally process results between steps.

agents¶: List of agents to execute in order

process_between_steps¶: Whether to process between agent calls

continue_on_error¶: Whether to continue if an agent fails

return_all_outputs¶: Return all intermediate outputs vs just final

Examples

Simple pipeline:

# Create pipeline: Researcher -> Analyst -> Writer
pipeline = SequentialAgent(
    name="report_pipeline",
    agents=[
        ResearchAgent(name="researcher"),
        AnalystAgent(name="analyst"),
        WriterAgent(name="writer")
    ]
)

result = pipeline.run("Create report on AI trends")
# Researcher finds data -> Analyst processes -> Writer creates report

With intermediate processing:

pipeline = SequentialAgent(
    name="enhanced_pipeline",
    agents=[research_agent, analysis_agent],
    process_between_steps=True,
    system_message="Enhance and clarify outputs between steps"
)

# Coordinator enhances outputs between each step

add_agent(agent)¶

Add an agent to the end of the sequence.

Parameters:: agent (haive.agents.simple.enhanced_simple_real.EnhancedAgentBase) – Agent to add to pipeline
Return type:: None

build_graph()¶

Build sequential execution graph.

Return type:: haive.core.graph.state_graph.base_graph2.BaseGraph

async execute_sequence(input_data)¶

Execute agents in sequence.

Parameters:: input_data (Any) – Initial input for the pipeline
Returns:: Final output or list of all outputs
Return type:: Any | list[Any]

get_pipeline_description()¶

Get human-readable pipeline description.

Return type:: str

insert_agent(index, agent)¶

Insert an agent at specific position.

Parameters:

index (int) – Position to insert at
agent (haive.agents.simple.enhanced_simple_real.EnhancedAgentBase) – Agent to insert

Return type:

None

remove_agent(index)¶

Remove agent at index.

Parameters:: index (int) – Position to remove from
Returns:: Removed agent or None
Return type:: haive.agents.simple.enhanced_simple_real.EnhancedAgentBase | None

setup_agent()¶

Setup sequential coordinator.

Return type:: None

classmethod validate_agents(v)¶

Validate agent list.

Parameters:: v (list[haive.agents.simple.enhanced_simple_real.EnhancedAgentBase])
Return type:: list[haive.agents.simple.enhanced_simple_real.EnhancedAgentBase]