# n8n Document Processing Pipeline with Supabase Vector DB

This n8n workflow automates the processing of documents uploaded to Google Drive, extracts their content, splits them into manageable chunks, generates OpenAI embeddings, and stores them in a Supabase Vector Store. This pipeline is designed to build a multi-client agentic RAG (Retrieval Augmented Generation) system for document processing.

## What it does

This workflow streamlines the process of ingesting documents for a RAG system by performing the following steps:

1.  **Monitors Google Drive**: It listens for new files added to a specified Google Drive folder.
2.  **Filters by Document Type**: It checks the file type and proceeds only if it's a document (e.g., PDF, DOCX, TXT).
3.  **Extracts File Content**: For each new document, it extracts the raw text content.
4.  **Splits Text into Chunks**: The extracted text is then divided into smaller, more manageable chunks using a Character Text Splitter, making it suitable for embedding.
5.  **Generates Embeddings**: Each text chunk is sent to OpenAI to generate vector embeddings.
6.  **Stores in Supabase Vector Store**: The text chunks along with their generated embeddings are stored in a Supabase Vector Store, making them ready for retrieval in a RAG application.
7.  **Logs Processing Status**: It updates a Postgres database with the status of the document processing (e.g., "processed").

## Prerequisites/Requirements

To use this workflow, you will need:

*   **n8n Instance**: A running n8n instance.
*   **Google Drive Account**: Configured with a credential in n8n to monitor for new files.
*   **OpenAI API Key**: Configured as a credential in n8n for generating embeddings.
*   **Supabase Project**: With a configured vector store and credentials in n8n.
*   **Postgres Database**: Configured with a credential in n8n to log document processing status.
*   **Langchain Nodes**: Ensure the `@n8n/n8n-nodes-langchain` package is installed in your n8n instance.

## Setup/Usage

1.  **Import the Workflow**: Download the JSON provided and import it into your n8n instance.
2.  **Configure Credentials**:
    *   **Google Drive Trigger**: Set up your Google Drive credential and specify the folder to monitor for new documents.
    *   **OpenAI Embeddings**: Configure your OpenAI API Key credential.
    *   **Supabase Vector Store**: Set up your Supabase project credentials (URL and API Key).
    *   **Postgres**: Configure your Postgres database credentials.
3.  **Activate the Workflow**: Once all credentials are set and configurations are in place, activate the workflow.

The workflow will now automatically process any new documents uploaded to your specified Google Drive folder, preparing them for use in your agentic RAG application.