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

# Creating a pipeline

> Create a pipeline and complete setup in the current tabbed pipeline workspace.

Pipeline creation starts with a small form and continues on the pipeline detail
page. There is no canvas or node editor.

## Create the pipeline

1. Open **Data Pipelines**.
2. Click **New Pipeline**.
3. Enter a name and optional description.
4. Choose the source connection.
5. Create and open the pipeline.

## Configure Source

1. Open **Source**.
2. Test the connection and click **Discover catalog**.
3. Enable the required streams.
4. Choose **Full table** or **Incremental** for each supported stream.
5. Enter a cursor for Incremental streams.
6. Inspect fields, preview rows, and click **Save stream settings**.

Start with one stream for the first run.

## Add a destination

1. Open **Destinations**.
2. Click **Add destination**.
3. Choose a destination connection and set its schema/database or Airtable
   base/table.
4. Click **Save destination**.

## Create a transformation

1. Open **Transformations** and select the destination.
2. Click **New transformation**.
3. Choose input streams, enter the output model name, and write SQL against the
   staged source relations.
4. **Save draft**, **Validate draft**, **Preview output**, and **Publish revision**.

Published revisions remain isolated to their owning destination.

## Configure output targets

1. Return to **Destinations** and open the destination.
2. Under **Published output targets**, enter each final table and Upsert key.
3. For Airtable, map output columns to writable fields and select merge fields.
4. Use **Create or verify destination table** for database targets when needed.
5. Save the destination.

## Validate and run

Open **Overview**, click **Validate pipeline**, and complete all readiness
checks. Then click **Run all** in the header and inspect the result in **Runs**.

## Add automation later

Open **Settings** after a successful manual run. Configure a fixed 1–24 hour
interval or a five-field UTC cron schedule, save it, and keep the pipeline
active.