Integrate Postgres Chat Memory with 500+ apps and services

This is the first version of a template for a RAG/GenAI App using WordPress content. As creating, sharing, and improving templates brings me joy 😄, feel free to reach out on LinkedIn if you have any ideas to enhance this template! How It Works This template includes three workflows: Workflow 1**: Generate embeddings for your WordPress posts and pages, then store them in the Supabase vector store. Workflow 2**: Handle upserts for WordPress content when edits are made. Workflow 3**: Enable chat functionality by performing Retrieval-Augmented Generation (RAG) on the embedded documents. Why use this template? This template can be applied to various use cases: Build a GenAI application that requires embedded documents from your website's content. Embed or create a chatbot page on your website to enhance user experience as visitors search for information. Gain insights into the types of questions visitors are asking on your website. Simplify content management by asking the AI for related content ideas or checking if similar content already exists. Useful for internal linking. Prerequisites Access to Supabase for storing embeddings. Basic knowledge of Postgres and pgvector. A WordPress website with content to be embedded. An OpenAI API key Ensure that your n8n workflow, Supabase instance, and WordPress website are set to the same timezone (or use GMT) for consistency. Workflow 1 : Initial Embedding This workflow retrieves your WordPress pages and posts, generates embeddings from the content, and stores them in Supabase using pgvector. Step 0 : Create Supabase tables Nodes : Postgres - Create Documents Table: This table is structured to support OpenAI embedding models with 1536 dimensions Postgres - Create Workflow Execution History Table These two nodes create tables in Supabase: The documents table, which stores embeddings of your website content. The n8n_website_embedding_histories table, which logs workflow executions for efficient management of upserts. This table tracks the workflow execution ID and execution timestamp. Step 1 : Retrieve and Merge WordPress Pages and Posts Nodes : WordPress - Get All Posts WordPress - Get All Pages Merge WordPress Posts and Pages These three nodes retrieve all content and metadata from your posts and pages and merge them. Important: ** **Apply filters to avoid generating embeddings for all site content. Step 2 : Set Fields, Apply Filter, and Transform HTML to Markdown Nodes : Set Fields Filter - Only Published & Unprotected Content HTML to Markdown These three nodes prepare the content for embedding by: Setting up the necessary fields for content embeddings and document metadata. Filtering to include only published and unprotected content (protected=false), ensuring private or unpublished content is excluded from your GenAI application. Converting HTML to Markdown, which enhances performance and relevance in Retrieval-Augmented Generation (RAG) by optimizing document embeddings. Step 3: Generate Embeddings, Store Documents in Supabase, and Log Workflow Execution Nodes: Supabase Vector Store Sub-nodes: Embeddings OpenAI Default Data Loader Token Splitter Aggregate Supabase - Store Workflow Execution This step involves generating embeddings for the content and storing it in Supabase, followed by logging the workflow execution details. Generate Embeddings: The Embeddings OpenAI node generates vector embeddings for the content. Load Data: The Default Data Loader prepares the content for embedding storage. The metadata stored includes the content title, publication date, modification date, URL, and ID, which is essential for managing upserts. ⚠️ Important Note : Be cautious not to store any sensitive information in metadata fields, as this information will be accessible to the AI and may appear in user-facing answers. Token Management: The Token Splitter ensures that content is segmented into manageable sizes to comply with token limits. Aggregate: Ensure the last node is run only for 1 item. Store Execution Details: The Supabase - Store Workflow Execution node saves the workflow execution ID and timestamp, enabling tracking of when each content update was processed. This setup ensures that content embeddings are stored in Supabase for use in downstream applications, while workflow execution details are logged for consistency and version tracking. This workflow should be executed only once for the initial embedding. Workflow 2, described below, will handle all future upserts, ensuring that new or updated content is embedded as needed. Workflow 2: Handle document upserts Content on a website follows a lifecycle—it may be updated, new content might be added, or, at times, content may be deleted. In this first version of the template, the upsert workflow manages: Newly added content** Updated content** Step 1: Retrieve WordPress Content with Regular CRON Nodes: CRON - Every 30 Seconds Postgres - Get Last Workflow Execution WordPress - Get Posts Modified After Last Workflow Execution WordPress - Get Pages Modified After Last Workflow Execution Merge Retrieved WordPress Posts and Pages A CRON job (set to run every 30 seconds in this template, but you can adjust it as needed) initiates the workflow. A Postgres SQL query on the n8n_website_embedding_histories table retrieves the timestamp of the latest workflow execution. Next, the HTTP nodes use the WordPress API (update the example URL in the template with your own website’s URL and add your WordPress credentials) to request all posts and pages modified after the last workflow execution date. This process captures both newly added and recently updated content. The retrieved content is then merged for further processing. Step 2 : Set fields, use filter Nodes : Set fields2 Filter - Only published and unprotected content The same that Step 2 in Workflow 1, except that HTML To Makrdown is used in further Step. Step 3: Loop Over Items to Identify and Route Updated vs. Newly Added Content Here, I initially aimed to use 'update documents' instead of the delete + insert approach, but encountered challenges, especially with updating both content and metadata columns together. Any help or suggestions are welcome! :) Nodes: Loop Over Items Postgres - Filter on Existing Documents Switch Route existing_documents (if documents with matching IDs are found in metadata): Supabase - Delete Row if Document Exists: Removes any existing entry for the document, preparing for an update. Aggregate2: Used to aggregate documents on Supabase with ID to ensure that Set Fields3 is executed only once for each WordPress content to avoid duplicate execution. Set Fields3: Sets fields required for embedding updates. Route new_documents (if no matching documents are found with IDs in metadata): Set Fields4: Configures fields for embedding newly added content. In this step, a loop processes each item, directing it based on whether the document already exists. The Aggregate2 node acts as a control to ensure Set Fields3 runs only once per WordPress content, effectively avoiding duplicate execution and optimizing the update process. Step 4 : HTML to Markdown, Supabase Vector Store, Update Workflow Execution Table The HTML to Markdown node mirrors Workflow 1 - Step 2. Refer to that section for a detailed explanation on how HTML content is converted to Markdown for improved embedding performance and relevance. Following this, the content is stored in the Supabase vector store to manage embeddings efficiently. Lastly, the workflow execution table is updated. These nodes mirros the **Workflow 1 - Step 3 nodes. Workflow 3 : An example of GenAI App with Wordpress Content : Chatbot to be embed on your website Step 1: Retrieve Supabase Documents, Aggregate, and Set Fields After a Chat Input Nodes: When Chat Message Received Supabase - Retrieve Documents from Chat Input Embeddings OpenAI1 Aggregate Documents Set Fields When a user sends a message to the chat, the prompt (user question) is sent to the Supabase vector store retriever. The RPC function match_documents (created in Workflow 1 - Step 0) retrieves documents relevant to the user’s question, enabling a more accurate and relevant response. In this step: The Supabase vector store retriever fetches documents that match the user’s question, including metadata. The Aggregate Documents node consolidates the retrieved data. Finally, Set Fields organizes the data to create a more readable input for the AI agent. Directly using the AI agent without these nodes would prevent metadata from being sent to the language model (LLM), but metadata is essential for enhancing the context and accuracy of the AI’s response. By including metadata, the AI’s answers can reference relevant document details, making the interaction more informative. Step 2: Call AI Agent, Respond to User, and Store Chat Conversation History Nodes: AI Agent** Sub-nodes: OpenAI Chat Model Postgres Chat Memories Respond to Webhook** This step involves calling the AI agent to generate an answer, responding to the user, and storing the conversation history. The model used is gpt4-o-mini, chosen for its cost-efficiency.

Edit 19/11/2024: As explained on the workflow, the AI Agent with the original system prompt was not effective when using gpt4-o-mini. To address this, I optimized the prompt to work better with this model. You can find the prompts I’ve tested on this Notion Page. And yes, there is one that works well with gpt4-o-mini. AI Agent to chat with you Search Console Data, using OpenAI and Postgres This AI Agent enables you to interact with your Search Console data through a chat interface. Each node is documented within the template, providing sufficient information for setup and usage. You will also need to configure Search Console OAuth credentials. Follow this n8n documentation to set up the OAuth credentials. Important Notes Correctly Configure Scopes for Search Console API Calls It’s essential to configure the scopes correctly in your Google Search Console API OAuth2 credentials. Incorrect configuration can cause issues with the refresh token, requiring frequent reconnections. Below is the configuration I use to avoid constant re-authentication: Of course, you'll need to add your client_id and client_secret from the Google Cloud Platform app you created to access your Search Console data. Configure Authentication for the Webhook Since the webhook will be publicly accessible, don’t forget to set up authentication. I’ve used Basic Auth, but feel free to choose the method that best meets your security requirements. 🤩💖 Example of awesome things you can do with this AI Agent

Video Guide I prepared a detailed guide to help you set up your workflow effectively, enabling you to extract insights from YouTube for content generation using an AI agent. Youtube Link Who is this for? This workflow is ideal for content creators, marketers, and analysts looking to enhance their YouTube strategies through data-driven insights. It’s particularly beneficial for individuals wanting to understand audience preferences and improve their video content. What problem does this workflow solve? Navigating the content generation and optimization process can be complex, especially without significant audience insight. This workflow automates insights extraction from YouTube videos and comments, empowering users to create more engaging and relevant content effectively. What this workflow does The workflow integrates various APIs to gather insights from YouTube videos, enabling automated commentary analysis, video transcription, and thumbnail evaluation. The main functionalities include: Extracting user preferences from comments. Transcribing video content for enhanced understanding. Analyzing thumbnails via AI for maximum viewer engagement insights. AI Insights Extraction: Automatically pulls comments and metrics from selected YouTube creators to evaluate trends and gaps. Dynamic Video Planning: Uses transcriptions to help creators outline video scripts and topics based on audience interest. Thumbnail Assessment: Provides analysis on thumbnail designs to improve click-through rates and viewer attraction. Setup N8N Workflow API Setup: Create a Google Cloud project and enable the YouTube Data API. Generate an API key to be included in your workflow requests. YouTube Creator and Video Selection: Start by defining a request to identify top creators based on their video views. Capture the YouTube video IDs for further analysis of comments and other video metrics. Comment Analysis: Gather comments associated with the selected videos and analyze them for user insights. Video Transcription: Utilize the insights from transcriptions to formulate content plans. Thumbnail Analysis: Evaluate your video thumbnails by submitting the URL through the OpenAI API to gain insights into their effectiveness.

Workflow Description Your workflow is an intelligent chatbot, using ++OpenAI assistant++, integrated with a backend that supports WhatsApp Business, designed to handle various use cases such as sales and customer support. Below is a breakdown of its functionality and key components: Workflow Structure and Functionality Chat Input (Chat Trigger) The flow starts by receiving messages from customers via WhatsApp Business. Collects basic information, such as session_id, to organize interactions. Condition Check (If Node) Checks if additional customer data (e.g., name, age, dependents) is sent along with the message. If additional data is present, a customized prompt is generated, which includes this information. The prompt specifies that this data is for the assistant's awareness and doesn’t require a response. Data Preparation (Edit Fields Nodes) Formats customer data and the interaction details to be processed by the AI assistant. Compiles the customer data and their query into a single text block. AI Responses (OpenAI Nodes) The assistant’s prompt is carefully designed to guide the AI in providing accurate and relevant responses based on the customer’s query and data provided. Prompts describe the available functionalities, including which APIs to call and their specific purposes, helping to prevent “hallucinated” or irrelevant responses. Memory and Context (Postgres Chat Memory) Stores context and messages in continuous sessions using a database, ensuring the chatbot maintains conversation history. API Calls The workflow allows the use of APIs with any endpoints you choose, depending on your specific use case. This flexibility enables integration with various services tailored to your needs. The OpenAI assistant understands JSON structures, and you can define in the prompt how the responses should be formatted. This allows you to structure responses neatly for the client, ensuring clarity and professionalism. Make sure to describe the purpose of each endpoint in the assistant’s prompt to help guide the AI and prevent misinterpretation. Customer Response Delivery After processing and querying APIs, the generated response is sent to the backend and ultimately delivered to the customer through WhatsApp Business. Best Practices Implemented Preventing Hallucinations** Every API has a clear description in its prompt, ensuring the AI understands its intended use case. Versatile Functionality** The chatbot is modular and flexible, capable of handling both sales and general customer inquiries. Context Persistence** By utilizing persistent memory, the flow maintains continuous interaction context, which is crucial for longer conversations or follow-up queries. Additional Recommendations Include practical examples in the assistant’s prompt, such as frequently asked questions or decision-making flows based on API calls. Ensure all responses align with the customer’s objectives (e.g., making a purchase or resolving technical queries). Log interactions in detail for future analysis and workflow optimization. This workflow provides a solid foundation for a robust and multifunctional virtual assistant 🚀

An intelligent chatbot that assists employees by answering common HR or IT questions, supporting both text and audio messages. This unique feature ensures employees can conveniently ask questions via voice messages, which are transcribed and processed just like text queries. How It Works Message Capture: When an employee sends a message to the chatbot in WhatsApp or Telegram (text or audio), the chatbot captures the input. Audio Transcription: For audio messages, the chatbot transcribes the content into text using an AI-powered transcription service (e.g., Whisper, Google Cloud Speech-to-Text). Query Processing: The transcribed text (or directly entered text) is sent to an AI service (e.g., OpenAI) to generate embeddings. These embeddings are used to search a vector database (e.g., Supabase or Qdrant) containing the company’s internal HR and IT documentation. The most relevant data is retrieved and sent back to the AI service to compose a concise and helpful response. Response Delivery: The chatbot sends the final response back to the employee, whether the input was text or audio. Set Up Steps Estimated Time**: 20–25 minutes Prerequisites**: Create an account with an AI provider (e.g., OpenAI). Connect WhatsApp or Telegram credentials in n8n. Set up a transcription service (e.g., Whisper or Google Cloud Speech-to-Text). Configure a vector database (e.g., Supabase or Qdrant) and add your internal HR and IT documentation. Import the workflow template into n8n and update environment variables for your credentials.