autogen.agentchat.group.patterns.pattern.Pattern

Pattern

Pattern(
    initial_agent: ConversableAgent,
    agents: list['ConversableAgent'],
    user_agent: ForwardRef('ConversableAgent') | None = None,
    group_manager_args: dict[str, Any] | None = None,
    context_variables: ContextVariables | None = None,
    group_after_work: TransitionTarget | None = None,
    exclude_transit_message: bool = True,
    summary_method: str | Callable[..., Any] | None = 'last_msg'
)

Base abstract class for all orchestration patterns.
Patterns provide a reusable way to define how agents interact within a group chat.
Each pattern encapsulates the logic for setting up agents, configuring handoffs, and determining the flow of conversation.
This is an abstract base class and should not be instantiated directly.
Use one of the concrete pattern implementations like AutoPattern, RoundRobinPattern, RandomPattern, or ManualPattern.
Initialize the pattern with the required components.

Parameters:

Name	Description
initial_agent	Type: ConversableAgent
agents	Type: list[‘ConversableAgent’]
user_agent	Type: ForwardRef(‘ConversableAgent’) | None Default: None
group_manager_args	Type: dict[str, typing.Any] | None Default: None
context_variables	Type: ContextVariables | None Default: None
group_after_work	Type: TransitionTarget | None Default: None
exclude_transit_message	Type: bool Default: True
summary_method	Type: str | Callable[…, Any] | None Default: ‘last_msg’

Static Methods

create_default

create_default(
    initial_agent: ConversableAgent,
    agents: list['ConversableAgent'],
    user_agent: ForwardRef('ConversableAgent') | None = None,
    group_manager_args: dict[str, Any] | None = None,
    context_variables: ContextVariables | None = None,
    exclude_transit_message: bool = True,
    summary_method: str | Callable[..., Any] | None = 'last_msg'
) -> 

Create a default pattern with minimal configuration.
This replaces the need for a separate BasePattern class by providing a factory method that creates a simple DefaultPattern instance.

Parameters:

Name	Description
initial_agent	The first agent to speak in the group chat. Type: ConversableAgent
agents	List of all agents participating in the chat. Type: list[‘ConversableAgent’]
user_agent	Optional user proxy agent. Type: ForwardRef(‘ConversableAgent’) | None Default: None
group_manager_args	Optional arguments for the GroupChatManager. Type: dict[str, typing.Any] | None Default: None
context_variables	Initial context variables for the chat. Type: ContextVariables | None Default: None
exclude_transit_message	Whether to exclude transit messages from the conversation. Type: bool Default: True
summary_method	Method for summarizing the conversation. Type: str | Callable[…, Any] | None Default: ‘last_msg’

Instance Methods

prepare_group_chat

prepare_group_chat(
    self,
    max_rounds: int,
    messages: list[dict[str, Any]] | str
) -> Tuple[list['ConversableAgent'], list['ConversableAgent'], ConversableAgent | None, ContextVariables, ConversableAgent, TransitionTarget, GroupToolExecutor, GroupChat, GroupChatManager, list[dict[str, Any]], ConversableAgent, list[str], list['Agent']]

Prepare the group chat for orchestration.
This is the main method called by initiate_group_chat to set up the pattern.
Subclasses must implement or extend this method to define pattern-specific behavior.

Parameters:

Name	Description
max_rounds	Maximum number of conversation rounds. Type: int
messages	Initial message(s) to start the conversation. Type: list[dict[str, typing.Any]] | str

Returns:

Type

Description

Tuple[list[‘ConversableAgent’], list[‘ConversableAgent’], ConversableAgent | None, ContextVariables, ConversableAgent, TransitionTarget, GroupToolExecutor, GroupChat, GroupChatManager, list[dict[str, Any]], ConversableAgent, list[str], list[‘Agent’]]

Tuple containing: - List of agents involved in the group chat - List of wrapped agents - User agent, if applicable - Context variables for the group chat - Initial agent for the group chat - Group-level after work transition for the group chat - Tool executor for the group chat - GroupChat instance - GroupChatManager instance - Processed messages - Last agent to speak - List of group agent names - List of temporary user agents