AgentOS Demo

Share one PostgresDb across AgentOS and its agent as the default database, with PgVector knowledge and the Agno MCP server.

Authentication is optional. Set OS_SECURITY_KEY to enable it for this AgentOS.

agentos_default_db.py

"""
AgentOS Demo

Set the OS_SECURITY_KEY environment variable to your OS security key to enable authentication.

Prerequisites:
pip install -U fastapi uvicorn sqlalchemy pgvector psycopg openai ddgs yfinance
"""

from agno.agent import Agent
from agno.db.postgres import PostgresDb
from agno.knowledge.knowledge import Knowledge
from agno.models.openai import OpenAIChat
from agno.os import AgentOS
from agno.tools.mcp import MCPTools
from agno.vectordb.pgvector import PgVector

# ---------------------------------------------------------------------------
# Create Example
# ---------------------------------------------------------------------------

# Database connection
db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"

# Create Postgres-backed memory store
db = PostgresDb(db_url=db_url)

# Create Postgres-backed vector store
vector_db = PgVector(
    db_url=db_url,
    table_name="agno_docs",
)
knowledge = Knowledge(
    name="Agno Docs",
    contents_db=db,
    vector_db=vector_db,
)

# Create your agents
agno_agent = Agent(
    name="Agno Agent",
    model=OpenAIChat(id="gpt-4.1"),
    tools=[MCPTools(transport="streamable-http", url="https://docs.agno.com/mcp")],
    knowledge=knowledge,
    markdown=True,
)

# Create the AgentOS
agent_os = AgentOS(
    id="agentos-demo",
    agents=[agno_agent],
    db=db,  # This is the default database for AgentOS, the agno_agent will use this
)
app = agent_os.get_app()


# ---------------------------------------------------------------------------
# Run Example
# ---------------------------------------------------------------------------

if __name__ == "__main__":
    agent_os.serve(app="agentos_default_db:app", port=7777, reload=True)

Run the Example

1. Set up your virtual environment

   Mac

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

   uv venv --python 3.12
   .venv\Scripts\activate
   
2. Install dependencies

   uv pip install -U "agno[mcp,os]" "psycopg[binary]" openai pgvector
   
3. Export your OpenAI API key

   Mac/Linux

   export OPENAI_API_KEY="your_openai_api_key_here"
   

   $Env:OPENAI_API_KEY="your_openai_api_key_here"
   
4. Run PgVector

   macOS / Linux

   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
   

   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
   
5. Optional: enable authentication

   Set OS_SECURITY_KEY before running to require AgentOS authentication. Leave it unset to run without authentication.

6. Run the example

   Save the code above as agentos_default_db.py, then run:

   python agentos_default_db.py
   

Full source: cookbook/05_agent_os/dbs/agentos_default_db.py