agents.reasoning_and_critique.reflection.config

Configuration for the Reflection Agent.

Classes

ReflectionAgentConfig	Configuration for an agent that uses reflection to improve responses.
ReflectionConfig	Configuration specific to the reflection mechanism.

Module Contents

class agents.reasoning_and_critique.reflection.config.ReflectionAgentConfig

Bases: haive.agents.simple.config.SimpleAgentConfig

Configuration for an agent that uses reflection to improve responses.

classmethod from_aug_llm(aug_llm, name=None, system_prompt=None, **kwargs)

Create a ReflectionAgentConfig from an existing AugLLMConfig.

Parameters:

• aug_llm (haive.core.engine.aug_llm.AugLLMConfig)

• name (str | None)

• system_prompt (str | None)

Return type:

ReflectionAgentConfig

classmethod from_scratch(system_prompt=None, model='gpt-4o', temperature=0.7, name=None, **kwargs)

Create a ReflectionAgentConfig from scratch.

Parameters:

• system_prompt (str | None)

• model (str)

• temperature (float)

• name (str | None)

Return type:

ReflectionAgentConfig

class agents.reasoning_and_critique.reflection.config.ReflectionConfig(/, **data)

Bases: pydantic.BaseModel

Configuration specific to the reflection mechanism.

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)