agents.planning.p_and_e.engines

AugLLM configurations for Plan and Execute Agent System.

This module defines the AugLLM configurations for planning, execution, and replanning agents.

Functions

create_executor_aug_llm_config([model_name, tools, ...])	Create AugLLM configuration for the execution agent.
create_planner_aug_llm_config([model_name, ...])	Create AugLLM configuration for the planning agent.
create_replan_aug_llm_config([model_name, ...])	Create AugLLM configuration for the replanning agent.

Module Contents

agents.planning.p_and_e.engines.create_executor_aug_llm_config(model_name='gpt-4o', tools=None, force_tool_use=False)

Create AugLLM configuration for the execution agent.

Parameters:

• model_name (str) – The LLM model to use

• temperature – Temperature for generation

• tools (list[langchain_core.tools.BaseTool] | None) – List of tools available to the executor

• force_tool_use (bool) – Whether to force tool usage

Returns:

Configured AugLLMConfig for execution

Return type:

haive.core.engine.aug_llm.AugLLMConfig

agents.planning.p_and_e.engines.create_planner_aug_llm_config(model_name='gpt-4o', temperature=0.1, use_context=True)

Create AugLLM configuration for the planning agent.

Parameters:

• model_name (str) – The LLM model to use

• temperature (float) – Temperature for generation (lower = more focused)

• use_context (bool) – Whether to use the context-aware prompt template

Returns:

Configured AugLLMConfig for planning

Return type:

haive.core.engine.aug_llm.AugLLMConfig

agents.planning.p_and_e.engines.create_replan_aug_llm_config(model_name='gpt-4o', temperature=0.2, max_replan_attempts=3)

Create AugLLM configuration for the replanning agent.

Parameters:

• model_name (str) – The LLM model to use

• temperature (float) – Temperature for generation

• max_replan_attempts (int) – Maximum number of replanning attempts

Returns:

Configured AugLLMConfig for replanning

Return type:

haive.core.engine.aug_llm.AugLLMConfig