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

# Pipelines API

> Endpoints for managing pipelines

# Pipelines API

<Warning>
  These endpoints require Session Authentication (user login cookie) and cannot be called using a `rag_` API key. They are used internally by the Jabrod dashboard.
</Warning>

## List Pipelines

`GET /api/rag/pipelines`

Returns a paginated list of pipelines belonging to the authenticated user.

**Query Parameters:**

* `page` (number): Page number (default: 1)
* `pageSize` (number): Items per page (default: 10)
* `search` (string): Search by pipeline name
* `status` (string): Filter by status

## Create Pipeline

`POST /api/rag/pipelines`

Creates a new RAG pipeline configuration. Enforces tier limits (e.g., Max 1 pipeline on Free tier).

**Body:**

* `name` (string)
* `description` (string)
* `chunkingStrategy` (string)
* `chunkSize` (number)
* `chunkOverlap` (number)
* `embeddingModel` (string)
* `vectorProvider` (string)
* `retrievalTopK` (number)

## Get Pipeline

`GET /api/rag/pipelines/[id]`

Returns the pipeline details, including a list of its associated Data Sources.

## Update Pipeline

`PATCH /api/rag/pipelines/[id]`

Updates pipeline configuration. Note: Updating chunking or embedding settings does *not* automatically re-index existing data sources. You must trigger a re-index manually.

## Delete Pipeline

`DELETE /api/rag/pipelines/[id]`

Deletes the pipeline. This triggers a cascade delete of all associated Data Sources and API Keys.