Google Drive and HTTP Request integration

The "Voice RAG Chatbot with ElevenLabs and OpenAI" workflow in n8n is designed to create an interactive voice-based chatbot system that leverages both text and voice inputs for providing information. Ideal for shops, commercial activities and restaurants How it works: Here's how it operates: Webhook Activation: The process begins when a user interacts with the voice agent set up on ElevenLabs, triggering a webhook in n8n. This webhook sends a question from the user to the AI Agent node. AI Agent Processing: Upon receiving the query, the AI Agent node processes the input using predefined prompts and tools. It extracts relevant information from the knowledge base stored within the Qdrant vector database. Knowledge Base Retrieval: The Vector Store Tool node interfaces with the Qdrant Vector Store to retrieve pertinent documents or data segments matching the user’s query. Text Generation: Using the retrieved information, the OpenAI Chat Model generates a coherent response tailored to the user’s question. Response Delivery: The generated response is sent back through another webhook to ElevenLabs, where it is converted into speech and delivered audibly to the user. Continuous Interaction: For ongoing conversations, the Window Buffer Memory ensures context retention by maintaining a history of interactions, enhancing the conversational flow. Set up steps: To configure this workflow effectively, follow these detailed setup instructions: ElevenLabs Agent Creation: Begin by creating an agent on ElevenLabs (e.g., named 'test_n8n'). Customize the first message and define the system prompt specific to your use case, such as portraying a character like a waiter at "Pizzeria da Michele". Add a Webhook tool labeled 'test_chatbot_elevenlabs' configured to receive questions via POST requests. Qdrant Collection Initialization: Utilize the HTTP Request nodes ('Create collection' and 'Refresh collection') to initialize and clear existing collections in Qdrant. Ensure you update placeholders QDRANTURL and COLLECTION accordingly. Document Vectorization: Use Google Drive integration to fetch documents from a designated folder. These documents are then downloaded and processed for embedding. Employ the Embeddings OpenAI node to generate embeddings for the downloaded files before storing them into Qdrant via the Qdrant Vector Store node. AI Agent Configuration: Define the system prompt for the AI Agent node which guides its behavior and responses based on the nature of queries expected (e.g., product details, troubleshooting tips). Link necessary models and tools including OpenAI language models and memory buffers to enhance interaction quality. Testing Workflow: Execute test runs of the entire workflow by clicking 'Test workflow' in n8n alongside initiating tests on the ElevenLabs side to confirm all components interact seamlessly. Monitor logs and outputs closely during testing phases to ensure accurate data flow between systems. Integration with Website: Finally, integrate the chatbot widget onto your business website replacing placeholder AGENT_ID with the actual identifier created earlier on ElevenLabs. By adhering to these comprehensive guidelines, users can successfully deploy a sophisticated voice-driven chatbot capable of delivering precise answers utilizing advanced retrieval-augmented generation techniques powered by OpenAI and ElevenLabs technologies.

How it Works This workflow automates the handling of incoming emails, summarizes their content, generates appropriate responses using a retrieval-augmented generation (RAG) approach, and obtains approval or suggestions before sending replies. Below is an explanation of its functionality divided into two main sections: Email Handling and Summarization: The process begins with the Email Trigger (IMAP) node which listens for new emails in a specified inbox. Once an email is received, the Markdown node converts its HTML content into plain text if necessary, followed by the Email Summarization Chain that uses AI to create a concise summary of up to 100 words. Response Generation and Approval: A Write email node generates a professional response based on the summarized content, ensuring brevity and professionalism while keeping within the word limit. Before sending out any automated replies, the system sends these drafts via Gmail for human review and approval through the Gmail node configured with free-text response options. If approved, the finalized email is sent back to the original sender using the Send Email node; otherwise, it loops back for further edits or manual intervention. Additionally, there's a Text Classifier node designed to categorize feedback from humans as either "Approved" or "Declined", guiding whether the email should proceed directly to being sent or require additional editing. Set Up Steps To replicate this workflow within your own n8n environment, follow these essential configuration steps: Configuration: Begin by setting up an n8n instance either locally or via cloud services offered directly from their official site. Import the provided JSON configuration file into your workspace, making sure all required credentials such as IMAP, SMTP, OpenAI API keys, etc., are properly set up under Credentials since multiple nodes rely heavily on external integrations for functionalities like reading emails, generating summaries, crafting replies, and managing approvals. Customization: Adjust parameters according to specific business needs, including but not limited to adjusting the conditions used during conditional checks performed by nodes like Approve?. Modify the template messages given to AI models so they align closely with organizational tone & style preferences while maintaining professionalism expected in business communications. Ensure correct mappings between fields when appending data to external systems where records might need tracking post-interaction completion, such as Google Sheets or similar platforms.

The provided workflow in n8n is designed to create a Business WhatsApp AI RAG (Retrieval-Augmented Generation) Chatbot. How it works: Webhook Setup: The workflow begins by setting up webhooks for verification and response. The Verify webhook receives GET requests and sends back a verification code, while the Respond webhook handles incoming POST requests from Meta regarding WhatsApp messages. Message Handling: Once a message is received, the workflow checks if the incoming JSON contains a user message. If it does, the message is processed further; otherwise, a generic response is sent. AI Agent Interaction: The user's message is passed to the AI Agent node, which uses a conversational agent with a predefined system message tailored for an electronics store. This ensures that the AI provides accurate and professional responses based on the knowledge base. Knowledge Base Utilization: The AI Agent references a knowledge base stored in Qdrant, a vector database. Documents from Google Drive are downloaded, vectorized using OpenAI embeddings, and stored in Qdrant for retrieval during conversations. Response Generation: The AI Agent generates a response using the OpenAI chat model (gpt-4o-mini) and sends it back to the user via WhatsApp. Set up steps: Create Qdrant Collection: Update the QDRANTURL and COLLECTION variables in the workflow. Use the Create collection HTTP request node to initialize the collection in Qdrant. Vectorize Documents: Configure the Get folder and Download Files nodes to fetch documents from a specified Google Drive folder. Use the Embeddings OpenAI node to generate embeddings for the downloaded files. Store the vectorized documents in Qdrant using the Qdrant Vector Store node. Configure Webhooks: Ensure both Verify and Respond webhooks have the same URL. Set the Verify webhook to use the GET HTTP method and the Respond webhook to use the POST HTTP method. Set Up AI Agent: Define the system prompt for the AI Agent, specifying guidelines for product information, technical support, customer service, and knowledge base usage. Link the AI Agent to the OpenAI chat model and configure any additional tools as needed. Test Workflow: Trigger the workflow manually using the When clicking ‘Test workflow’ node to ensure all components are functioning correctly. Monitor the flow of data through the nodes and verify that responses are being generated and sent accurately. By following these steps, the workflow will be fully operational, enabling a robust AI-powered chatbot capable of handling customer inquiries via WhatsApp.

This workflow is ideal for businesses looking to automate their email responses, especially for handling inquiries about company information. It leverages AI to ensure accurate and professional communication. How It Works Email Trigger: The workflow starts with the Email Trigger (IMAP) node, which monitors an email inbox for new messages. When a new email arrives, it triggers the workflow. Email Preprocessing: The Markdown node converts the email's HTML content into plain text for easier processing by the AI models. Email Summarization: The Email Summarization Chain node uses an AI model (DeepSeek R1) to generate a concise summary of the email. The summary is limited to 100 words and is written in Italian. Email Classification: The Email Classifier node categorizes the email into predefined categories (e.g., "Company info request"). If the email does not fit any category, it is classified as "other". Email Response Generation: The Write email node uses an AI model (OpenAI) to draft a professional response to the email. The response is based on the email content and is limited to 100 words. The Review email node uses another AI model (DeepSeek) to review and format the drafted response. It ensures the response is professional and formatted in HTML (e.g., using , , , tags where necessary). Email Sending: The Send Email node sends the reviewed and formatted response back to the original sender. Vector Database Integration: The Qdrant Vector Store node retrieves relevant information from a vector database (Qdrant) to assist in generating accurate responses. This is particularly useful for emails classified as "Company info request". The Embeddings OpenAI node generates embeddings for the email content, which are used to query the vector database. Document Vectorization: The workflow includes steps to create and refresh a Qdrant collection (Create collection and Refresh collection nodes). Documents from Google Drive are downloaded (Get folder and Download Files nodes), processed into embeddings (Embeddings OpenAI1 node), and stored in the Qdrant vector store (Qdrant Vector Store1 node). Set Up Steps Configure Email Trigger: Set up the Email Trigger (IMAP) node with the appropriate IMAP credentials to monitor the email inbox. Set Up AI Models: Configure the DeepSeek R1, OpenAI, and DeepSeek nodes with the appropriate API credentials for text summarization, response generation, and review. Set Up Email Classification: Define the categories in the Email Classifier node (e.g., "Company info request", "Other"). Ensure the OpenAI 4-o-mini node is configured to assist in classification. Set Up Vector Database: Configure the Qdrant Vector Store and Qdrant Vector Store1 nodes with the appropriate Qdrant API credentials and collection details. Set up the Embeddings OpenAI and Embeddings OpenAI1 nodes to generate embeddings for the email content and documents. Set Up Document Processing: Configure the Get folder and Download Files nodes to access and download documents from Google Drive. Use the Token Splitter and Default Data Loader nodes to process and split the documents into manageable chunks for vectorization. Set Up Email Sending: Configure the Send Email node with the appropriate SMTP credentials to send responses. Test the Workflow: Trigger the workflow manually using the When clicking ‘Test workflow’ node to ensure all steps execute correctly. Verify that emails are summarized, classified, and responded to accurately. Activate the Workflow: Once tested, activate the workflow to automate the process of handling incoming emails. Key Features Automated Email Handling**: Automatically processes incoming emails, summarizes them, and generates professional responses. AI-Powered Classification**: Uses AI to classify emails into relevant categories for targeted responses. Vector Database Integration**: Retrieves relevant information from a vector database to enhance response accuracy. Document Vectorization**: Processes and stores documents from Google Drive in a vector database for quick retrieval. Professional Email Formatting**: Ensures responses are professionally formatted and concise.

This AI Agent is designed to streamline the process of creating realistic images of clothing products worn by models, eliminating the need for expensive photoshoots. The primary goal is to automate the generation of virtual try-on images for an eCommerce store selling clothing, using advanced image processing technologies. Example of results How It Works Triggering the Workflow: The workflow can be triggered manually using the When clicking ‘Test workflow’ node or automatically via the Schedule Trigger node, which runs the workflow at regular intervals (e.g., every 5 minutes). Data Retrieval: The Get new product node retrieves data from a Google Sheets document containing the URLs of the model image, the clothing product image, and the WooCommerce product ID. The document also includes a column for the resulting virtual try-on image URL, which is initially empty. Setting Up the Request: The Set data node prepares the data for the AI request by assigning the model image URL and the product image URL to variables. AI Image Generation: The Create Image node sends a request to API to generate a virtual try-on image. The request includes the URLs of the model and the clothing product. When you register for the API service you will get 1$ for free. For continuous work add API credits to your account. Image Retrieval: The Get Url image node retrieves the URL of the generated virtual try-on image. The Get File image node downloads the generated image from the provided URL. Image Storage: The Upload Image node uploads the generated image to a specified Google Drive folder for storage. Updating Google Sheets: The Update result node updates the Google Sheets document with the URL of the generated virtual try-on image. Updating WooCommerce: The Update product node updates the corresponding product in WooCommerce by adding the generated virtual try-on image to the product's image gallery. Functionality This AI Agent is designed to streamline the process of creating realistic images of clothing products worn by models, eliminating the need for expensive photoshoots. The primary goal is to automate the generation of virtual try-on images for an eCommerce store selling clothing, using advanced image processing technologies. By offering realistic images of clothing items worn by models, this automation process saves time and resources, making product catalog management more efficient. This approach not only enhances the competitiveness of the eCommerce store but also provides greater flexibility in creating high-quality visual content adaptable to various digital marketing campaigns. Key Features Automated Virtual Try-On**: Generates realistic images of models wearing clothing items using AI. Google Sheets Integration**: Retrieves and updates data in Google Sheets for seamless data management. Image Storage**: Uploads generated images to Google Drive for easy access and archiving. WooCommerce Integration**: Updates product pages with virtual try-on images, enhancing the shopping experience. Scheduled Automation**: Can be triggered manually or run at regular intervals for continuous processing. This workflow is ideal for eCommerce businesses looking to enhance their product catalogs with high-quality, realistic images without the need for costly photoshoots. It leverages AI to provide a cost-effective and efficient solution for virtual try-on image generation.

This workflow combines OpenAI, Retrieval-Augmented Generation (RAG), and WooCommerce to create an intelligent personal shopping assistant. It handles two scenarios: Product Search: Extracts user intent (keywords, price ranges, SKUs) and fetches matching products from WooCommerce. General Inquiries: Answers store-related questions (e.g., opening hours, policies) using RAG and documents stored in Google Drive. How It Works Chat Interaction & Intent Detection Chat Trigger**: Starts when a user sends a message ("When chat message received"). Information Extractor**: Uses OpenAI to analyze the message and determine if the user is searching for a product or asking a general question. Extracts: search (true/false). keyword, priceRange, SKU, category (if product-related). Example: { "search": true, "keyword": "red handbags", "priceRange": { "min": 50, "max": 100 }, "SKU": "BAG123", "category": "women's accessories" } Product Search (WooCommerce Integration) AI Agent**: If search: true, routes the request to the personal_shopper tool. WooCommerce Node: Queries the WooCommerce store using extracted parameters (keyword, priceRange, SKU). Filters products in stock (stockStatus: "instock"). Returns matching products (e.g., "red handbags under €100"). General Inquiries (RAG System) RAG Tool**: If search: false, uses the Qdrant Vector Store to retrieve store information from documents. Google Drive Integration: Documents (e.g., store policies, FAQs) are stored in Google Drive. Downloaded, split into chunks, and embedded into Qdrant for semantic search. OpenAI Chat Model: Generates answers based on retrieved documents (e.g., "Our store opens at 9 AM"). Set Up Steps Configure the RAG System Google Drive Setup**: Upload store documents . Update the Google Drive2 node with your folder ID. Qdrant Vector Database**: Clean the collection (update Qdrant Vector Store node with your URL). Use Embeddings OpenAI to convert documents into vectors. Configure OpenAI & WooCommerce OpenAI Credentials**: Add your API key to all OpenAI nodes (OpenAI Chat Model, Embeddings OpenAI, etc.). WooCommerce Integration**: Connect your WooCommerce store (credentials in the personal_shopper node). Ensure product data is synced and accessible. Customize the AI Agent Intent Detection**: Modify the Information Extractor’s system prompt to align with your store’s terminology. RAG Responses**: Update the tool description to reflect your store’s documents. Notes This template is ideal for e-commerce businesses needing a hybrid assistant for product discovery and customer support.