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

# Quickstart — Python SDK

> Connect your Celery workers to Sluice in under 2 minutes with the Python SDK.

By the end of this guide, your Celery jobs will appear in the Sluice dashboard — with real-time state tracking, search, and management actions.

## Prerequisites

* Python 3.11 or later
* Celery 5.3 or later with a Redis broker
* A [Sluice account](https://sluice.sh/signup) (free, no credit card required)

## 1. Create a connection

Sign in to the [Sluice dashboard](https://sluice.sh) and go to **Connections**. Click **Add Connection**, pick "Python SDK", and name your connection (e.g., "Production Celery"). You'll receive:

* An **API key** — starts with `sk_`
* A **Connection ID** — a UUID that identifies this connection

Copy both values.

## 2. Install the SDK

```bash theme={null}
pip install sluice-celery
```

## 3. Initialize Sluice

Add an import and an `init()` call to your Celery configuration — wherever you define your Celery app:

<CodeGroup>
  ```python celeryconfig.py theme={null}
  import sluice

  sluice.init(
      api_key="sk_live_...",
      connection_id="550e8400-e29b-41d4-a716-446655440000",
  )

  # Your existing Celery config below
  broker_url = "redis://localhost:6379/0"
  ```

  ```python Django settings.py theme={null}
  # settings.py
  import sluice

  sluice.init(
      api_key="sk_live_...",
      connection_id="550e8400-e29b-41d4-a716-446655440000",
  )

  # Or use environment variables (recommended for production):
  # SLUICE_API_KEY=sk_live_...
  # SLUICE_CONNECTION_ID=550e8400-e29b-41d4-a716-446655440000
  # Then just call: sluice.init()
  ```

  ```python FastAPI / Flask theme={null}
  # In your app's entrypoint (main.py, app.py, etc.)
  import sluice

  sluice.init(
      api_key="sk_live_...",
      connection_id="550e8400-e29b-41d4-a716-446655440000",
  )

  # Your existing Celery + web framework code unchanged
  ```
</CodeGroup>

<Tip>
  **Use environment variables in production.** Set `SLUICE_API_KEY` and `SLUICE_CONNECTION_ID` as env vars, then call `sluice.init()` with no arguments. This keeps secrets out of your codebase.
</Tip>

## 4. Restart your worker

```bash theme={null}
celery -A your_app worker -l info
```

You should see Sluice confirm the connection in the worker logs:

```
[sluice] Django project detected.
[sluice] SDK initialized successfully.
[sluice] Celery worker 'celery@your-hostname' configured for monitoring.
```

## 5. See your data

Open the [Sluice dashboard](https://sluice.sh). Send a task from your application and watch it appear in real time.

That's it — you're monitoring Celery.

## What the SDK does automatically

When `sluice.init()` runs, the SDK:

1. **Auto-configures Celery events** — sets `worker_send_task_events=True`, `task_send_sent_event=True`, and `task_track_started=True`. These are all `False` by default in Celery, and without them, most monitoring data is invisible.
2. **Installs a Celery Bootstep** — hooks into the worker lifecycle to capture task events as they happen.
3. **Normalizes and forwards events** — batches events and sends them to the Sluice API over HTTPS.
4. **Never crashes your worker** — all runtime errors are caught and logged. The SDK operates under a strict "production cannot fail" philosophy.

<Info>
  The SDK auto-detects whether you're running Django or standalone Celery and configures itself accordingly. No framework-specific setup is needed.
</Info>

## Next steps

* [SDK Reference](/sdk/reference) — all `sluice.init()` parameters and environment variables
* [Safety Guarantees](/sdk/safety) — how the SDK protects your production workers
* [Troubleshooting](/troubleshooting/no-data) — if jobs aren't appearing in the dashboard