Team with Memory Manager

The local database setup below requires Docker.

Set update_memory_on_run=True and pass a custom MemoryManager to run it on every team run. Automatic extraction can create, update, and delete user memories based on the latest input. The manager's delete_memories flag does not disable deletion on this automatic path; it controls agentic memory updates.

Code

team_with_memory_manager.py

from uuid import uuid4

from agno.agent import Agent
from agno.db.postgres import PostgresDb
from agno.memory import MemoryManager
from agno.models.openai import OpenAIResponses
from agno.team import Team
from rich.pretty import pprint

db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"
db = PostgresDb(db_url=db_url)

session_id = str(uuid4())
john_doe_id = "john_doe@example.com"

memory_manager = MemoryManager(model=OpenAIResponses(id="gpt-5.2"))

agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
)

team = Team(
    model=OpenAIResponses(id="gpt-5.2"),
    memory_manager=memory_manager,
    members=[agent],
    db=db,
    update_memory_on_run=True,
)

team.print_response(
    "My name is John Doe and I like to hike in the mountains on weekends.",
    stream=True,
    user_id=john_doe_id,
    session_id=session_id,
)

team.print_response(
    "What are my hobbies?", stream=True, user_id=john_doe_id, session_id=session_id
)

memories = team.get_user_memories(user_id=john_doe_id)
print("John Doe's memories:")
pprint(memories)

Usage

Set up your virtual environment

uv venv --python 3.12
source .venv/bin/activate

Install dependencies

uv pip install -U agno openai sqlalchemy 'psycopg[binary]'

Run PgVector

docker run -d \
  -e POSTGRES_DB=ai \
  -e POSTGRES_USER=ai \
  -e POSTGRES_PASSWORD=ai \
  -e PGDATA=/var/lib/postgresql \
  -v pgvolume:/var/lib/postgresql \
  -p 5532:5432 \
  --name pgvector \
  agnohq/pgvector:18

Set environment variables

export OPENAI_API_KEY=your_openai_api_key_here

Run the example

python team_with_memory_manager.py