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

# start • AI Agents Framework

> start: Start scheduled agent execution.

# start

<div className="flex items-center gap-2">
  <Badge color="purple">Method</Badge>
</div>

> This is a method of the [**AgentScheduler**](../classes/AgentScheduler) class in the [**agent\_scheduler**](../modules/agent_scheduler) module.

Start scheduled agent execution.

## Signature

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
def start(schedule_expr: str, max_retries: int, run_immediately: bool) -> bool
```

## Parameters

<ParamField query="schedule_expr" type="str" required={true}>
  Schedule expression (e.g., "hourly", "\*/1h", "3600")
</ParamField>

<ParamField query="max_retries" type="int" required={false} default="3">
  Maximum retry attempts on failure
</ParamField>

<ParamField query="run_immediately" type="bool" required={false} default="False">
  If True, run agent immediately before starting schedule
</ParamField>

### Returns

<ResponseField name="Returns" type="bool">
  True if scheduler started successfully
</ResponseField>

## Uses

* `logger.warning`
* `ScheduleParser.parse`
* `logger.info`
* `threading.Thread`
* `start`
* `logger.error`

## Used By

* [`PraisonAgentExecutor.execute`](../functions/PraisonAgentExecutor-execute)
* [`AgentScheduler.start`](../functions/AgentScheduler-start)
* [`Profiler.streaming`](../functions/Profiler-streaming)
* [`Profiler.streaming_async`](../functions/Profiler-streaming_async)
* [`Profiler.memory`](../functions/Profiler-memory)

## Source

<Card title="View on GitHub" icon="github" href="https://github.com/MervinPraison/PraisonAI/blob/main/src/praisonai/praisonai/agent_scheduler.py#L169">
  `praisonai/agent_scheduler.py` at line 169
</Card>