Workflow Templates

📩 Gmail → GPT → Supabase | Task Extractor This n8n workflow automates the extraction of actionable tasks from unread Gmail messages using OpenAI's GPT API, stores the resulting task metadata in Supabase, and avoids re-processing previously handled emails. ✅ What It Does Triggers on a schedule to check for unread emails in your Gmail inbox. Loops through each email individually using SplitInBatches. Checks Supabase to see if the email has already been processed. If it's a new email: Formats the email content into a structured GPT prompt Calls ChatGPT-4o to extract structured task data Inserts the result into your emails table in Supabase 🧰 Prerequisites Before using this workflow, you must have: An active n8n Cloud or self-hosted instance A connected Gmail account with OAuth credentials in n8n A Supabase project with an emails table and: ALTER TABLE emails ADD CONSTRAINT unique_email_id UNIQUE (email_id); An OpenAI API key with access to GPT-4o or GPT-3.5-turbo 🔐 Required Credentials | Name | Type | Description | |-----------------|------------|-----------------------------------| | Gmail OAuth | Gmail | To pull unread messages | | OpenAI API Key | OpenAI | To generate task summaries | | Supabase API | HTTP | For inserting rows via REST API | 🔁 Environment Variables or Replacements Supabase_TaskManagement_URI → e.g., https://your-project.supabase.co Supabase_TaskManagement_ANON_KEY → Your Supabase anon key These are used in the HTTP request to Supabase. ⏰ Scheduling / Trigger Triggered using a Schedule node Default: every X minutes (adjust to your preference) Uses a Gmail API filter: unread emails with label = INBOX 🧠 Intended Use Case > Designed for productivity-minded professionals who want to extract, summarize, and store actionable tasks from incoming email — without processing the same email twice or wasting GPT API credits. This is part of a larger system integrating GPT, calendar scheduling, and optional task platforms (like ClickUp). 📦 Output (Stored in Supabase) Each processed email includes: email_id subject sender received_at body (email snippet) gpt_summary (structured task) requires_deep_work (from GPT logic) deleted (initially false)

n8n Workflow: AI-Powered Stock Chart Analysis Bot for Telegram This is a powerful n8n automation workflow that integrates a Telegram bot with OpenAI's multimodal large language model (GPT-4 Vision) to provide users with real-time stock chart analysis. Workflow Breakdown Receive Image:** The workflow is initiated by a Telegram Trigger. It activates whenever a user sends an image (e.g., a stock's candlestick chart) to a designated Telegram chat, automatically downloading the file. Image Pre-processing:** To optimize the AI's performance and efficiency, the Edit Image node resizes the incoming image to a standard 512x512 pixel format. AI Vision Analysis:** The processed image is then passed to a LangChain Chain, which utilizes the OpenAI GPT-4 Vision model. A sophisticated system prompt instructs the AI to act as a professional stock analyst. Intelligent Interpretation:** The AI analyzes the image to identify the stock's name, price trend (uptrend, downtrend, or sideways), key support/resistance levels, and volume changes. It then generates a comprehensive analysis report combining technical indicators and market sentiment. Structured Output:** To ensure reliability and consistency, the AI's output is parsed into a specific JSON format. This structure includes a search_word (for the industry/sector) and the main content (the analysis text). Send Response:** Finally, the workflow extracts the content field from the JSON output and uses the Telegram node to send this professional analysis back to the user as a text message in the same chat. Key Features User-Friendly:** Users simply send an image to get an analysis, requiring no complex commands. Instant & Efficient:** The entire analysis and response process is fully automated and completed within moments. Professional-Grade Analysis:** Leverages the advanced image recognition and reasoning capabilities of GPT-4 Vision to deliver insights comparable to those of a human analyst. Reliable & Consistent:** The use of structured output ensures that the format of the response is always consistent and easy to read or process further.

🏖️ AI-Based Tour Itineraries via Email Using OpenAI & Pinecone Vector Search Overview This workflow automates the process of handling new tour package requests received via email, analyzes the request, and provides personalized tour package recommendations using AI and a vector database. It’s designed to streamline customer interactions and deliver quick, relevant responses. Precondition Create a Embedded Tour Package Database (refer to the link below): Pinecone Database setup Register and create API Keys for OpenAI, Pinecone Database. Copy Mail Credentials to access Email Inbox from n8n node This workflow automates the process of extracting tour information from PDF files stored in a Google Drive folder, processes and vectorizes the extracted data, and stores it in a Pinecone vector database for efficient querying. This is especially useful for building AI-powered search or recommendation systems for travel packages. 🛠️ Tools & Nodes Used Email Trigger (IMAP): Monitors the inbox for new tour package requests. Text Classifier: Categorizes incoming emails (e.g., New Request, Follow-up, Other). Code Node: Extracts and structures relevant data from the email (subject, sender, content, etc.). Tour Recommendation AI Agent: An AI agent that interprets the request and formulates a prompt for package recommendations. OpenAI & OpenRouter Chat Models: Used for natural language understanding and generating responses. Simple Memory: Maintains context for ongoing conversations. Pinecone Vector Store: Stores and retrieves tour packages using semantic search. Embeddings (OpenAI): Converts text data into vector embeddings for similarity search. Answer Questions with a Vector Store: Retrieves the most relevant packages from Pinecone. Send Email: Sends the AI-generated recommendations back to the customer. 🔄 Process & Flow Email Reception: The workflow starts with the Email Trigger (IMAP) node, which listens for new emails in the inbox. Classification: The Text Classifier node determines if the email is a new tour package request. Data Extraction: The Code node parses the email, extracting key details like sender, subject, and content. AI Agent Processing: The Tour Recommendation AI Agent receives the structured request and crafts a prompt for package recommendations. Vector Search: The agent queries the Pinecone Vector Store, which holds previously created tour packages, using OpenAI embeddings for semantic matching. Recommendation Generation: The AI agent selects the top 3 most relevant packages and generates a friendly, personalized response. Response Delivery: The Send Email node sends the recommendations back to the customer. 🚀 Recommendations & Improvements for Next Version Error Handling: Add error handling nodes to manage failed email parsing or AI response issues. Logging & Analytics: Integrate logging to track requests, recommendations, and customer responses for continuous improvement. Personalization: Enhance the AI agent to consider customer history or preferences for even more tailored recommendations. Multi-language Support: Add language detection and translation for international customers. Feedback Loop: Include a mechanism for customers to rate recommendations, feeding this data back into the system for improved future suggestions. Attachment Handling: Enable the workflow to process attachments (e.g., customer itineraries or preferences). Scalability: Consider batching or queueing requests if email volume increases. 💡 Conclusion This workflow demonstrates how n8n, combined with AI and vector databases, can automate and personalize customer service in the travel industry. With a few enhancements, it can become even more robust and customer-centric!

This is a workflow that might come handy after using loops. They usually leave you with items spread across different "runs". The Code node in this example workflow merges them into a single run, so you have a single list of items which is often easier to work with. Simply adjust the node name inside the Code node as needed. The idea is based on this older workflow template.

This workflow generates CSV files containing a list of 10 random users with specific characteristics using OpenAI's GPT-4 model. It then splits this data into batches, converts it to CSV format, and saves it to disk for further use. The execution of the workflow begins from here when triggered manually. "OpenAI" Node. This uses the OpenAI API to generate random user data. The input to the OpenAI API is a fixed string, which asks for a list of 10 random users with some specific attributes. The attributes include a name and surname starting with the same letter, a subscription status, and a subscription date (if they are subscribed). There is also a short example of the JSON object structure. This technique is called one-shot prompting. "Split In Batches" Node. This node is used to handle the OpenAI responses one by one. "Parse JSON" Node. This node converts the content of the message received from the OpenAI node (which is in string format) into a JSON object. "Make JSON Table" Node. This node is used to convert the JSON data into a tabular format, which is easier to handle for further data processing. "Convert to CSV" Node. This node converts the table format data received from the "Make JSON Table" node into CSV format and assigns a file name. "Save to Disk" Node. This node is used to save the CSV generated in the previous node to disk in the ".n8n" directory. The workflow is designed in a circular manner. So, after saving the file to disk, it goes back to the "Split In Batches" node to process the OpenAI output, until all batches are processed.

Transforming Emails into Podcasts 🎙️ Check out this channel for example. The n8n workflow described here aims to revolutionize the way users engage with promotional emails by converting them into entertaining audio podcasts. This innovative project leverages automation through n8n to streamline tasks and enhance user experience. Project Benefit 🎧🌟 The primary goal of this project is to transform "CATEGORY_PROMOTIONS" emails into engaging audio content. By converting text into speech, users can enjoy promotional material hands-free, making it easier to consume information while on the go or relaxing. The workflow consists of several key steps orchestrated seamlessly to deliver a delightful experience to users. How to Use the Workflow: Gmail trigger Node: Initiates the workflow by fetching "CATEGORY_PROMOTIONS" emails at regular intervals. The Gmail Trigger node in your N8N workflow is set to poll for new emails every minute and is configured to filter emails with the label "CATEGORY_PROMOTIONS" before triggering the workflow. Steps to Use Filters Inside the Gmail Trigger Node: Configure Gmail Trigger Node: Set "Poll Times" to "Every Minute" to check for new emails at regular intervals. Enable the "Simple" toggle if you want to simplify the node interface. Under "Filters", specify the label IDs you want to filter by. In this case, it's set to "CATEGORY_PROMOTIONS". Adjust any additional options as needed. // Configure Gmail Trigger node pollTimes: { item: [ { mode: "everyMinute" } ] }, simple: false, filters: { labelIds: [ "CATEGORY_PROMOTIONS" ] }, options: {} Save and Execute: Save your workflow and execute it to start monitoring your Gmail account for new emails with the specified label filter. By following these steps, your workflow will effectively trigger based on new emails that match the "CATEGORY_PROMOTIONS" label in your Gmail account. Get message content Node: Extracts the email content for processing. Summarization Chain Node: Generates concise summaries using advanced methods for better readability. Delete the unnecessary items Node: Removes irrelevant details from the email content. Text to Free TTS Node: Converts the summary text into speech using Free TTS technology. Convert from base64 to File Node: Transforms the audio data into a compatible file format. Merge Text with Audio Node: Combines the text and audio components seamlessly. Aggregate in same cell Node: Gathers all processed data for finalization. Send Message to Telegram Node: Dispatches the audio message along with a caption to a designated Telegram chat ID. By automating these tasks, the workflow ensures efficient communication and delivers content in a more engaging format, fostering a positive user experience. Configuration Instructions: The configuration of this workflow involves setting up the necessary nodes and establishing connections between them. Each node performs a specific function crucial to the overall operation of the workflow. Additionally, credentials need to be provided for accessing Gmail and OpenAI services to enable seamless data processing and summarization. Utilizing Text-to-Speech API 🎧 In addition to n8n automation, an external Text-to-Speech API plays a pivotal role in generating audio content from text data. By sending a POST request with JSON data containing the text and voice preferences, users can quickly receive audio files of the converted content. The API offers a straightforward interface for text-to-speech conversion, making it ideal for creating audio clips efficiently. To access this API, simply submit the desired text and voice selection to receive the generated speech audio file. The API endpoint can be accessed at https://tiktok-tts.weilnet.workers.dev/api/generation or through https://tiktokvoicegenerator.com/. In conclusion, this n8n workflow coupled with a Text-to-Speech API presents a powerful solution for transforming emails into captivating podcasts, enhancing user engagement and communication effectiveness. By embracing automation and innovative technologies, this project aims to improve user experience and streamline content delivery processes. 🌈✨🚀

OpenAI Assistant is a powerful tool, but at the time of writing it doesn't automatically remember past messages from a conversation. This workflow demonstrates how to get around this, by managing the chat history in n8n and passing it to the assistant when required. This makes it possible to use OpenAI Assistant for chatbot use cases. Note that to use this template, you need to be on n8n version 1.28.0 or later.

The Agent Decisioner is a dynamic, AI-powered routing system that automatically selects the most appropriate large language model (LLM) to respond to a user's query based on the query’s content and purpose. This workflow ensures dynamic, optimized AI responses by intelligently routing queries to the best-suited model. Advantages 🔁 Automatic Model Routing:** Automatically selects the best model for the job, improving efficiency and relevance of responses. 🎯 Optimized Use of Resources:** Avoids overuse of expensive models like GPT-4 by routing simpler queries to lightweight models. 📚 Model-Aware Reasoning:** Uses detailed metadata about model capabilities (e.g., reasoning, coding, web search) for intelligent selection. 📥 Modular and Extendable:** Easy to integrate with other tools or expand by adding more models or custom decision logic. 👨‍💻 Ideal for RAG and Multi-Agent Systems:** Can serve as the brain behind more complex agent frameworks or Retrieval-Augmented Generation pipelines. How It Works Chat Trigger: The workflow starts when a user sends a message, triggering the Routing Agent. Model Selection: The AI Agent analyzes the query and selects the best-suited model from the available options (e.g., Claude 3.7 Sonnet for coding, Perplexity/Sonar for web searches, GPT-4o Mini for reasoning). Structured Output: The agent returns a JSON response with the user’s prompt and the chosen model. Execution: The selected model processes the query and generates a response, ensuring optimal performance for the task. Set Up Steps Configure Nodes: Chat Trigger: Set up the webhook to receive user messages. Routing Agent (AI Agent): Define the system message with model strengths and JSON output rules. OpenRouter Chat Model: Connect to OpenRouter for model access. Structured Output Parser: Ensure it validates the JSON response format (prompt + model). Execution Agent (AI Agent1): Configure it to forward the prompt to the selected model. Connect Nodes: Link the Chat Trigger to the Routing Agent. Connect the OpenRouter Chat Model and Output Parser to the Routing Agent. Route the parsed JSON to the Execution Agent, which uses the chosen model via OpenRouter Chat Model1. Credentials: Ensure OpenRouter API credentials are correctly set for both chat model nodes. Test & Deploy: Activate the workflow and test with sample queries to verify model selection logic. Adjust the routing rules if needed for better accuracy. Need help customizing? Contact me for consulting and support or add me on Linkedin.

The 🌐🤖 AI Agent Chatbot with Jina.ai Webpage Scraper workflow is a powerful automation designed to integrate real-time web scraping capabilities into an AI-driven chatbot. Here's how it works and why it's important: How It Works 💬 Chat Trigger: The workflow begins when a user sends a chat message, triggering the "When chat message received" node. 🧠 AI Agent Processing: The input is passed to the "Jina.ai Web Scraping Agent," which uses advanced AI logic to interpret the user’s query and determine the information needed. 🌐 Web Scraping: The agent utilizes the "HTTP Request" node to scrape real-time data from a user-provided URL, enabling the chatbot to fetch and analyze live website content. 🗂️ Memory Management: The "Window Buffer Memory" node ensures context retention by storing and managing conversational history, allowing for seamless interactions. 🤖 Language Model Integration: The scraped data is processed using the "gpt-4o-mini" language model, which generates clear, accurate, and contextually relevant responses for the user. Why It's Cool ⏱️ Real-Time Information Retrieval**: This workflow empowers users to access up-to-date web content directly through a chatbot, eliminating manual web searches. ✨ Enhanced User Experience**: By combining web scraping with conversational AI, it delivers precise answers tailored to user queries in real time. 🔄 Versatility**: It can be applied across various domains, such as customer support, research, or data analysis, making it a valuable tool for businesses and individuals alike. ⚙️ Automation Efficiency**: Automating web scraping and response generation saves time and effort while ensuring accuracy.

Dessix Moss Ttsd Text Generator Description MOSS-TTSD (text to spoken dialogue) is an open-source bilingual spoken dialogue synthesis model that supports both Chinese and English. It can transform dialogue scripts between two speakers into natural, expressive conversational speech. Overview This n8n workflow integrates with the Replicate API to use the dessix/moss-ttsd model. This powerful AI model can generate high-quality text content based on your inputs. Features Easy integration with Replicate API Automated status checking and result retrieval Support for all model parameters Error handling and retry logic Clean output formatting Parameters Optional Parameters seed** (integer, default: 42): Random seed for reproducibility text** (string, default: [S1]你好[S2]你好，最近怎么样[S1]还不错，你呢[S2]我也挺好的，谢谢关心): Dialogue text, format: [S1]Speaker 1 content[S2]Speaker 2 content[S1]... use_normalize** (boolean, default: True): Whether to use text normalization (recommended for better handling of numbers, punctuation, etc.) reference_text_speaker1** (string, default: 周一到周五每天早晨七点半到九点半的直播片段,言下之意呢就是废话有点多,大家也别嫌弃,因为这都是直播间最真实的状态了): Reference text for speaker 1 (corresponding to reference audio) reference_text_speaker2** (string, default: 如果大家想听到更丰富更及时的直播内容,记得在周一到周五准时进入直播间,和大家一起畅聊新消费新科技新趋势): Reference text for speaker 2 (corresponding to reference audio) reference_audio_speaker1** (string, default: None): Reference audio file for speaker 1 (optional, for voice cloning) reference_audio_speaker2** (string, default: None): 说话者2的参考音频文件（可选，用于声音克隆）/ Reference audio file for speaker 2 (optional, for voice cloning) How to Use Set up your Replicate API key in the workflow Configure the required parameters for your use case Run the workflow to generate text content Access the generated output from the final node API Reference Model: dessix/moss-ttsd API Endpoint: https://api.replicate.com/v1/predictions Requirements Replicate API key n8n instance Basic understanding of text generation parameters

Pizza Ordering Chatbot with OpenAI - Menu, Orders & Status Tracking Introduction This workflow template is designed to automate order processing for a pizza store using OpenAI and n8n. The chatbot acts as a virtual assistant to handle customer inquiries related to menu details, order placement, and order status tracking. Features The chatbot provides an interactive experience for customers by performing the following functions: Menu Inquiry: When a customer asks about the menu, the chatbot responds with a list of available pizzas, prices, and additional options. Order Placement: If a customer places an order, the chatbot confirms order details, provides a summary, informs the customer that the order is being processed, and expresses gratitude. Order Status Tracking: If a customer asks about their order status, the chatbot retrieves details such as order date, pizza type, and quantity, providing real-time updates. Prerequisites Before setting up the workflow, ensure you have the following: OpenAI account** (Sign up here) OpenAI API key** to interact with GPT-3.5 n8n instance** running locally or on a server (Installation Guide) Configuration Steps Step 1: Set Up OpenAI API Credentials Log in to OpenAI's website. Navigate to API Keys under your account settings. Click Create API Key and copy the key for later use. Step 2: Configure OpenAI Node in n8n Open n8n and create a new workflow. Click Add Node and search for OpenAI. Select OpenAI from the list. In the OpenAI node settings, click "Create New" under the Credentials section. Enter a name for the credentials (e.g., "PizzaBot OpenAI Key"). Paste your API Key into the field. Click Save. Step 3: Set Up the Chatbot Logic Connect the AI Agent Builder Node to the OpenAI Node and HTTP Request Node. Configure the OpenAI Node with the following settings: Model: gpt-3.5-turbo Prompt: Provide dynamic text based on customer inquiries (e.g., "List available pizzas," "Place an order for Margherita pizza," "Check my order status"). Temperature: Adjust based on desired creativity (recommended: 0.7). Max Tokens: Limit response length (recommended: 150). Add multiple HTTP Request Node: For Get Products: Fetch stored menu data and return details. For Order Product: Capture order details, generate an order ID, and confirm with the customer. For Get Order: Retrieve order details based on the order ID and display progress. Step 4: Testing and Deployment Click Execute Workflow to test the chatbot. Open the Chat Message node, then copy the chat URL to access the chatbot in your browser. Interact with the chatbot by asking different queries (e.g., "What pizzas do you have?" or "I want to order a Pepperoni pizza"). Verify responses and adjust prompts or configurations as needed. Deploy the workflow and integrate it with a messaging platform (e.g., Telegram, WhatsApp, or a website chatbot). Conclusion This n8n workflow enables a fully functional pizza ordering chatbot using OpenAI's GPT-3.5. Customers can view menus, place orders, and track their order status efficiently. You can further customize the chatbot by refining prompts, adding new features, or integrating with external databases for order management. 🚀 Happy automating!

Welcome to Ozki Your Data Analyst Agent V1. The Ozki Data Analyst Agent is designed to analyze data from Google Sheets. To use it, you'll need to provide the URL of your Google Sheet file. The agent will then process the data and provide you with analysis results, including key performance indicators (KPIs). Configuration: Configure your credentials on the OpenAI model or select the n8n free OpenAI credits. Set up your agent memory. Use Simple Memory as default. Set your credentials to Google Sheets. Log in with the Google Sheet tool. Instructions: Start with a "Hi" to get the instructions. Ozki needs your Google Sheet URL, which you can get from the address bar of your browser when you have the file open. Follow the conversation with Ozki for your data analysis results.