Skip to main content

​
Overview

A customer support agent that combines all learning stores to provide faster resolutions over time. Uses namespace isolation for multi-tenant deployments where different organizations share the system but have separate knowledge.

​
Code

support_agent.py
from agno.agent import Agent
from agno.db.postgres import PostgresDb
from agno.knowledge import Knowledge
from agno.knowledge.embedder.openai import OpenAIEmbedder
from agno.learn import (
    EntityMemoryConfig,
    LearnedKnowledgeConfig,
    LearningMachine,
    LearningMode,
    SessionContextConfig,
    UserProfileConfig,
)
from agno.models.openai import OpenAIResponses
from agno.vectordb.pgvector import PgVector, SearchType

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

# Shared knowledge base for solutions
knowledge = Knowledge(
    vector_db=PgVector(
        db_url=db_url,
        table_name="support_kb",
        search_type=SearchType.hybrid,
        embedder=OpenAIEmbedder(id="text-embedding-3-small"),
    ),
)


def create_support_agent(customer_id: str, ticket_id: str, org_id: str) -> Agent:
    """Create a support agent for a specific ticket."""
    return Agent(
        model=OpenAIResponses(id="gpt-5.2"),
        db=db,
        instructions=(
            "You are a helpful support agent. "
            "Check if similar issues have been solved before. "
            "Save successful solutions for future reference."
        ),
        learning=LearningMachine(
            knowledge=knowledge,
            user_profile=UserProfileConfig(mode=LearningMode.ALWAYS),
            session_context=SessionContextConfig(enable_planning=True),
            entity_memory=EntityMemoryConfig(
                mode=LearningMode.ALWAYS,
                namespace=f"org:{org_id}:support",
            ),
            learned_knowledge=LearnedKnowledgeConfig(mode=LearningMode.AGENTIC),
        ),
        user_id=customer_id,
        session_id=ticket_id,
        markdown=True,
    )


if __name__ == "__main__":
    org_id = "acme"

    # Ticket 1: First customer with login issue
    agent = create_support_agent("[email protected]", "ticket_001", org_id)
    agent.print_response(
        "I can't log into my account. It says 'invalid credentials' "
        "even though I know my password is correct. I'm using Chrome.",
        stream=True,
    )

    # Solution worked - agent saves it
    agent.print_response("Clearing the cache worked! Thanks!", stream=True)

    # Ticket 2: Similar issue - agent finds prior solution
    agent2 = create_support_agent("[email protected]", "ticket_002", org_id)
    agent2.print_response(
        "Login not working in Chrome, says wrong password but I'm sure it's right.",
        stream=True,
    )

​
How It Works

  • User Profile: Tracks customer history and preferences
  • Session Context: Manages current ticket with goal/plan/progress
  • Entity Memory: Stores product info, past tickets, shared across organization
  • Learned Knowledge: Saves successful solutions for future retrieval
Key pattern: The namespace=f"org:{org_id}:support" isolates entity memory per organization while the learned knowledge base can be shared globally or per-org.

​
Run This Example

git clone https://github.com/agno-agi/agno.git
cd agno/cookbook/08_learning

# Setup (requires Docker for Postgres)
./setup_venv.sh
./cookbook/scripts/run_pgvector.sh

# Run
python 07_patterns/support_agent.py
Full source: agno/cookbook/08_learning/07_patterns/support_agent.py