> ## Documentation Index
> Fetch the complete documentation index at: https://phidatainc.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Running the scheduler inside AgentOS with automatic polling

> Enable cron polling with scheduler=True on AgentOS and create schedules through the REST API.

```python scheduler_with_agentos.py theme={null}
"""Running the scheduler inside AgentOS with automatic polling.

This example demonstrates the primary DX for the scheduler:
- Setting scheduler=True on AgentOS to enable cron polling
- The poller starts automatically on app startup and stops on shutdown
- Schedules are created via the REST API (POST /schedules)
- The internal service token handles auth between scheduler and agent endpoints

Run with:
    .venvs/demo/bin/python cookbook/05_agent_os/scheduler/scheduler_with_agentos.py
"""

from agno.agent import Agent
from agno.db.sqlite import SqliteDb
from agno.models.openai import OpenAIChat
from agno.os import AgentOS

# --- Setup ---

db = SqliteDb(id="scheduler-os-demo", db_file="tmp/scheduler_os_demo.db")

greeter = Agent(
    name="Greeter",
    model=OpenAIChat(id="gpt-4o-mini"),
    instructions=["You are a friendly greeter."],
    db=db,
)

reporter = Agent(
    name="Reporter",
    model=OpenAIChat(id="gpt-4o-mini"),
    instructions=["You summarize news headlines in 2-3 sentences."],
    db=db,
)

# Create AgentOS with scheduler enabled.
# This does three things:
#   1. Registers the /schedules REST endpoints
#   2. Starts a SchedulePoller on app startup (polls every 15s by default)
#   3. Auto-generates an internal service token for scheduler -> agent auth
agent_os = AgentOS(
    name="Scheduled OS",
    agents=[greeter, reporter],
    db=db,
    scheduler=True,
    scheduler_poll_interval=15,  # seconds between poll cycles (default: 15)
    # scheduler_base_url="http://127.0.0.1:7777",  # default
    # internal_service_token="my-secret",  # auto-generated if omitted
)
app = agent_os.get_app()

# --- Run the server ---
# Once running, create schedules via:
#
#   curl -X POST http://127.0.0.1:7777/schedules \
#     -H "Content-Type: application/json" \
#     -d '{
#       "name": "greet-every-5-min",
#       "cron_expr": "*/5 * * * *",
#       "endpoint": "/agents/greeter/runs",
#       "payload": {"message": "Say hello!"}
#     }'
#
# The poller will pick it up on the next poll cycle and run the agent.

if __name__ == "__main__":
    agent_os.serve(app="scheduler_with_agentos:app", reload=True)
```

## Run the Example

<Steps>
  <Snippet file="create-venv-step.mdx" />

  <Step title="Install dependencies">
    ```bash theme={null}
    uv pip install -U "agno[os]" openai
    ```
  </Step>

  <Step title="Export your OpenAI API key">
    <CodeGroup>
      ```bash Mac/Linux theme={null}
      export OPENAI_API_KEY="your_openai_api_key_here"
      ```

      ```bash Windows theme={null}
      $Env:OPENAI_API_KEY="your_openai_api_key_here"
      ```
    </CodeGroup>
  </Step>

  <Step title="Run the example">
    Save the code above as `scheduler_with_agentos.py`, then run:

    ```bash theme={null}
    python scheduler_with_agentos.py
    ```
  </Step>
</Steps>

Full source: [cookbook/05\_agent\_os/scheduler/scheduler\_with\_agentos.py](https://github.com/agno-agi/agno/blob/main/cookbook/05_agent_os/scheduler/scheduler_with_agentos.py)