AI Email Auto-Responder System (AI RAG Agent for Email Inbox)

This n8n workflow sets up an intelligent AI-powered email auto-responder system. It leverages an AI agent with Retrieval Augmented Generation (RAG) capabilities to understand incoming emails, retrieve relevant information from a knowledge base (Google Drive/Docs), and craft appropriate responses. It can also manage tasks in Notion based on email content.

What it does

This workflow automates email responses and task management through the following steps:

1. Triggers on new emails: Listens for new emails in a specified Gmail inbox.
2. Filters emails (Implicit): Although not explicitly shown with an If node, the design suggests that the AI Agent will process emails to determine if a response or action is needed.
3. Processes email content with an AI Agent: An AI Agent (powered by an OpenAI Chat Model) analyzes the incoming email.
4. Retrieves information from a Vector Store: The AI Agent uses a "Vector Store Question Answer Tool" to query a Pinecone Vector Store, which holds embeddings of documents from Google Drive and Google Docs. This acts as the RAG component, providing context for generating responses.
5. Generates email responses: Based on the email content and retrieved information, the AI Agent formulates a response.
6. Sends automated email replies: Uses Gmail to send the AI-generated response.
7. Manages tasks in Notion (Optional/Tool-based): The AI Agent can also utilize a "Call n8n Workflow Tool" to interact with Notion, potentially creating or updating tasks based on email instructions or content.
8. Ingests and prepares knowledge base documents: A separate (sub)workflow, triggered by new Google Drive files or executed manually, processes documents from Google Drive and Google Docs:
   • Loads document content using a "Default Data Loader".
   • Splits the text into smaller chunks using a "Recursive Character Text Splitter".
   • Generates embeddings for these text chunks using "Embeddings OpenAI".
   • Stores these embeddings in a "Pinecone Vector Store" to build the RAG knowledge base.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance.
• Gmail Account: Configured as a credential in n8n for the Gmail Trigger and Gmail node.
• Google Drive/Docs Account: Configured as a credential in n8n for the Google Drive Trigger, Google Drive, and Google Docs nodes. This is where your knowledge base documents will reside.
• OpenAI API Key: Configured as a credential in n8n for the OpenAI Chat Model and Embeddings OpenAI nodes.
• Pinecone Account: Configured as a credential in n8n for the Pinecone Vector Store.
• Notion Account: Configured as a credential in n8n for the Notion node (if task management is desired).

Setup/Usage

1. Import the workflow: Download the provided JSON and import it into your n8n instance.
2. Configure Credentials:
   • Set up your Gmail, Google Drive, Google Docs, OpenAI, Pinecone, and Notion credentials in n8n.
   • Ensure the Google Drive Trigger is configured to monitor the correct folder(s) for your knowledge base documents.
   • Ensure the Gmail Trigger is configured for the inbox you want to monitor.
3. Build your Knowledge Base:
   • Place your relevant documents (e.g., FAQs, product information, support guides) in the Google Drive folder(s) monitored by the Google Drive Trigger.
   • Manually execute the "When Executed by Another Workflow" branch (or set up a schedule) to process existing documents and build the Pinecone Vector Store.
4. Activate the Workflow: Once all credentials are set and the knowledge base is built, activate the main workflow.

The system will now automatically listen for new emails, use AI to understand them, retrieve context from your documents, generate responses, and potentially manage tasks in Notion.