Teams

A Team is a collection of Agents (or other sub-teams) that work together to accomplish tasks. Teams can either “coordinate”, “collaborate” or “route” to solve a task. A Team has a list of members that can be instances of either Agent or Team.

from agno.team import Team
from agno.agent import Agent

team = Team(members=[
    Agent(name="Agent 1", role="You answer questions in English"),
    Agent(name="Agent 2", role="You answer questions in Chinese"),
    Team(name="Team 1", role="You answer questions in French"),
])

The team will transfer tasks to the members depending on the mode of the team. As with agents, teams support the following features:

Model: Set the model that is used by the “team leader” to delegate tasks to the team members.
Instructions: Instruct the team leader on how to solve problems. The names, descriptions and roles of team members are automatically provided to the team leader.
Tools: If the team leader needs to be able to use tools directly, you can add tools to the team.
Reasoning: Enables the team leader to “think” before responding or delegating tasks to team members, and “analyze” the results of team members’ responses.
Knowledge: If the team needs to search for information, you can add a knowledge base to the team. This is accessible to the team leader.
Storage: The Team’s session history and state is stored in a database. This enables your team to continue conversations from where they left off, enabling multi-turn, long-term conversations.
Memory: Gives Teams the ability to store and recall information from previous interactions, allowing them to learn user preferences and personalize their responses.