Agents & Automations

Every time a new image containing text is posted to your Slack channel, Make will detect the text using Google Cloud Vision (OCR) and send it as a new message to a selected Slack channel.

How it works Fetch transaction notification emails (including attachments). Clean up data. Let AI (Basic LLM Chain node) generate bookkeeping item. Send to Google Sheet. Details The example fetches email from Gmail labels, suggested using filters to automatically organize email into the labels. Data will be sent to the "raw data" sheet. Example Google Sheet:Link to Google Sheet

Deep web scrapper + RAG: Automation recursively downloads each page of the target website and extracts links, emails, texts, and PDF documents. Then, all extracted data goes into RAG, from which you can later extract data via chat or any other interface. Steps to follow: 1. Create a Supabase account and project. 2. Connect Supabase to n8n. 3. Connect PostgreSQL from Supabase to n8n. 4. Create Supabase tables and functions. 5. Run the automation. 6. If automation times out, you can re-run it with a click-to-start workflow node connected to the 'Check Supabase' node. 7. Sometimes, an HTTP request fails and causes automation to mark the URL as failed, but you can re-activate these URLs (after automation is finished) with another sub-flow. Then simply re-run the main web-scrapper automation.

Create blog posts in three languages, a Facebook update, and an email campaign with Claude and WordPress. This is all done with a simple email prompt. The template is part of this webinar.

Who this is for? Extract & Summarize Indeed Company Info is an automated workflow that extracts the Indeed company profile information using Bright Data Web Unlocker, transforms it using Google Gemini’s LLM, and forwards the transformed response with the summary to a specified webhook for downstream use. This workflow is tailored for: - Recruiters and HR teams looking to assess companies quickly during talent sourcing. - Job seekers researching potential employers and needing summarized company insights. - Market researchers and analysts monitoring competitor or industry players. What problem is this workflow solving? Searching and evaluating company profiles on Indeed manually can be time-consuming and inefficient, especially when dealing with large volumes of companies. Manually browsing, copying, and summarizing company descriptions, reviews, and ratings from Indeed hinders productivity and limits real-time insights. This workflow solves this by: - Automating the extraction of company details from Indeed using Bright Data Web Unlocker. - Summarizing the raw data using Google Gemini's language model for a quick, human-readable overview. - Sending the transformed response with the summary to a chosen endpoint, like Slack, Notion, Airtable, or a custom webhook. What this workflow does This automated pipeline does the following: - Scrape Indeed company profile pages (e.g., ratings, description, reviews) using Bright Data’s Web Unlocker. - Transform the scraped content into structured JSON using n8n’s built-in tools. - Summarize and extract meaningful insights using Google Gemini's large language model. - Forward the summarized data to a specified webhook or app for real-time access, storage, or analysis. How to customize this workflow to your needs This workflow is built to be flexible - whether you're a company or a market researcher, entrepreneur, or data analyst. Here’s how you can adapt it to fit your specific use case: - Changing the data source: Replace the Indeed search input with other job or business listing platforms if needed (e.g., Glassdoor, Crunchbase). - Refining the LLM prompt: Tailor the Gemini prompt to transform or summarize the Indeed company information in a specific format. - Routing the output to different destinations: Send summaries or transformed responses to Google Sheets, Airtable, or CRMs like HubSpot or Salesforce, etc.

This n8n template imports purchase order submissions from Outlook and converts attached purchase order forms in XLSX format into structured output. Data entry jobs with user-submitted XLSX forms are time-consuming, incredibly mundane but necessary tasks which are critical to business operations. While we could dream of system overhauls and modernization, the fact is that change is hard. There is another way, however - using n8n and AI! n8n offers an end-to-end solution to parse XLSX form attachments using LLM-powered OCR and send the extracted output to your ERP or otherwise. How it works An Outlook trigger is used to watch for incoming purchase order forms submitted via a shared inbox. The email attachment for the submission is a form in XLSX format, which is imported into the workflow. The 'Extract from File' node is used with the 'code' node to convert the XLSX file to markdown. This is so our LLM can understand it. The Information Extractor node is used to read and extract the relevant purchase order details and line items from the form. A simple validation step is used to check for common errors such as missing PO number or the amounts not matching up. A notification is automated to reply to the buyer if issues are found. Once validation passes, a confirmation is sent to the buyer and the purchase order structured output can be sent along to internal systems. How to use This template only works if you're expecting and receiving forms in XLSX format. These can be invoices, request forms, as well as purchase order forms. Update the Outlook nodes with your email or other emails as required. What's next? I've omitted the last steps to send to an ERP or accounting system as this is dependent on your organization. Customizing the workflow This template should work for other Excel files. Some will be more complicated than others, so experiment with different parsers and extraction tools and strategies. Customize the Information Extractor Schema to pull out the specific data you need. For example, capture any notes or comments given by the buyer.

Learn how to build a fully automated AI Avatar Social Media system that creates talking head AI clone videos, WITHOUT having to film or edit yourself. This tutorial combines n8n, AI tools agent, HeyGen, and Blotato to research, write, create, and distribute talking head AI clone videos to every social media platform every single day. 100% automated. • n8n - workflow automation that runs daily • AI agent - uses HackerNews and ChatGPT • HeyGen - create realistic AI clone/avatar • Blotato - publish to all social platforms

AI-Powered SEO Keyword Research Workflow with n8n This workflow automates comprehensive keyword research for content creation. This n8n workflow automates SEO keyword research using AI and data-driven analytics. It combines OpenAI's language models with DataForSEO's analytics to generate comprehensive keyword strategies for content creation. The workflow is triggered by a webhook from NocoDB, processes the input data through multiple stages, and returns a detailed content brief with optimized keywords. Workflow Architecture The workflow follows a structured process: - Input Collection: Receives data via webhook from NocoDB - Topic Expansion: Generates keywords using AI - Keyword Metrics Analysis: Gathers search volume, CPC, and difficulty metrics - Competitor Analysis: Analyzes competitor content for ranking keywords - Final Strategy Creation: Combines all data to generate a comprehensive keyword strategy - Output Storage: Saves results back to NocoDB and sends notifications NocoDB Integration The workflow integrates with two tables in NocoDB: - Input Table Schema: This table collects the input parameters for the keyword research. - Output Table Schema: This table stores the generated keyword strategy. Data Flow The workflow handles data in the following sequence: - Webhook Trigger: Receives input from NocoDB when a new keyword research request is created - Field Extraction: Extracts primary topic, competitor URLs, audience, and other parameters - AI Topic Expansion: Uses OpenAI to generate related keywords, categorized by type and intent - Keyword Analysis: Sends primary keywords to DataForSEO to get search volume, CPC, and difficulty - Competitor Research: Analyzes competitor pages to identify their keyword rankings - Strategy Generation: Combines all data to create a comprehensive keyword strategy - Storage & Notification: Saves the strategy to NocoDB and sends a notification to Slack Core Components 1. Topic Expansion: This component uses OpenAI and a structured output parser to generate: - 20 primary keywords - 30 long-tail keywords with search intent - 15 question-based keywords - 10 related topics 2. DataForSEO Integration: Two API endpoints are used: - Search Volume & CPC: Gets monthly search volume and cost-per-click data - Keyword Difficulty: Evaluates how difficult it would be to rank for each keyword 3. Competitor Analysis: This component: - Analyzes competitor URLs to identify which keywords they rank for - Identifies content gaps or opportunities - Determines the search intent their content targets 4. Final Keyword Strategy: The AI-generated strategy includes: - Top 10 primary keywords with metrics - 15 long-tail opportunities with low competition - 5 question-based keywords to address in content - Content structure recommendations - 3 potential content titles optimized for SEO Setup Requirements To use this workflow, you'll need: - n8n Instance: Either cloud or self-hosted - NocoDB Account: For data input and storage - API Keys: - OpenAI API key - DataForSEO API credentials - Slack API token (for notifications) - Database Setup: Create the required tables in NocoDB as described above Possible Improvements The workflow could be enhanced with the following improvements: - Enhanced Keyword Strategy: - Add topic clustering to group related keywords - Enhance the final output with more specific content structure suggestions - Include word count recommendations for each content section - Additional Data Sources: - Integrate Google Search Console data for existing content optimization - Add Google Trends data to identify rising topics - Include sentiment analysis for different keyword groups - Improved Competitor Analysis: - Analyze content length and structure from top-ranking pages - Identify common backlink sources for competitor content - Extract content headings to better understand content organization - Automation Enhancements: - Add scheduling capabilities to run updates on existing content - Implement content performance tracking over time - Create alert thresholds for changes in keyword difficulty or search volume Example Output Here is an example output generated based on the following inputs: - Primary Topic: AI Automation - Competitor URLs: n8n.io, zapier.com, make.com - Target Audience: Small Business Owners - Content Type: Landing Page - Location: United States - Language: English The workflow provides a powerful automation for content marketers and SEO specialists to develop data-driven keyword strategies with minimal manual effort.

Who is it for? This automation is for support project managers, which helps not only to keep developers informed but also automatically keep clients in the loop—especially useful if you are managing SLA-like agreements. It is actually a simple incident management board using a free Kanban board that is extended in functionality via N8N. How It Works? The script has two entry points. The first one is the incident form. When incident details are provided, automation gets incident definitions from the database and pushes both information to AI. AI compares definitions with client requests, refines incident priority, and pushes it into the NocoDB database. The second is a schedule trigger, which is responsible for regular notifications on task status. If a task is not picked up or delivered in a proper time, then emails or Slack messages are sent both to the client and the responsible developer. How to set up? 1. Clone the automation. 2. Create (samples below) two NocoDB tables: one with definitions and the second that serves as a Kanban board (mind column naming!). 3. Set up email and Slack connections. You should be ready to go! Different incident naming If your incident level naming is different, you need to update a few nodes and a few columns in NocoDB. This is because incident naming must be unified through: automation flow, incident definitions, and column NocoDB select fields. So be sure that the following is the same: - NocoDB: Incident definitions, column "Title" - NocoDB: Tasks table, single select fields: - "expected category" - "assigned category" - N8N: Incident Form "Incident Desired Category" NocoDB TablesIncident definitions table | Title | Definition | Response time | Resolution time | Default assignee | |------------------|----------------|---------------|------------------|------------------| | single line text | text | number | number | email | Tasks table | email | message | expected category | internal notes | assigned category | status | expected response | expected resolution | assignee | assignee slack | |-------|---------|-------------------|----------------|-------------------|--------|-------------------|---------------------|----------|-----------------| | email | text | single select | text | single select | single select | date and time | date and time | email | slack username | Use Kanban board Simply set up the Kanban view and stack by the "status" field. What's More? That's actually it. I hope that this automation will help your support line be much more streamlined! There is actually more that you could do with this automation, but it really depends on your needs. For example, you could add an Email trigger to handle incoming support requests (but remember to adjust nodes accordingly). Another thing is that you could create different notification schemas, depending on your needs (for example, you may want a day or two delay before notifying the client that a task is overdue).

👉 Build a Phone Agent to qualify outbound leads and schedule inbound calls. Who is this for? This workflow is designed for sales teams, call centers, and businesses handling both outbound and inbound lead calls who want to automate their qualification, follow-up, and call documentation process without manual intervention. It’s ideal for teams using Google Sheets, RetellAI, OpenAI, and Gmail as part of their tech stack. Real-World Use Cases 🛍 E-commerce – Instantly handle product FAQs and order status checks, 24/7. 🏬 Retail Stores – Share store hours, directions, and return policies without lifting a finger. 🍽 Restaurants – Take reservations or answer menu questions automatically. 💼 Service Providers – Book appointments or consultations while you focus on your craft. 📞 Any Local Business – Deliver friendly, consistent phone support — no live agent required. What problem is this workflow solving? Managing lead calls at scale can be chaotic—between scheduling outbound qualification calls, handling inbound appointment requests, and making sure every call is documented and followed up. This workflow automates the entire process, reducing human error and saving time by: ✅ Sending reminders to reps for outbound calls ✅ Automatically placing calls with RetellAI ✅ Handling inbound calls and checking caller details ✅ Generating and emailing call summaries automatically What this workflow does This n8n template connects Google Sheets, RetellAI, OpenAI, and Gmail into a seamless workflow: Outbound Lead Qualification Workflow - Triggers when a new lead is added to Google Sheets - Sends an SMS notification to remind the rep to call in 5 minutes - (Optional) Waits 5 minutes - Initiates an automated call to the lead via RetellAI Inbound Call Appointment Scheduler - Receives inbound calls from RetellAI (via webhook) - Checks if the caller’s number exists in Google Sheets - Responds to RetellAI with a success or error message Post-Call Workflow - Receives post-call data from RetellAI - Filters only analyzed calls - Updates the lead’s record in Google Sheets - Uses OpenAI to generate a call summary - Emails the summary to a team inbox or rep Setup ✅ You need an active RetellAI API key. Sign up for RetellAI, create an agent, and set the webhook URLs (n8n_call for call events). ✅ Your Google Sheet must have a column for phone numbers (e.g., "Phone"). ✅ Gmail account connected and authorized in n8n. ✅ OpenAI API key added to your environment variables or credentials. Configure your Google Sheets node with the correct spreadsheet ID and range. Add your RetellAI API key to the HTTP request nodes. Connect your Gmail account in the Gmail node. Add your OpenAI key in the OpenAI node. How to customize this workflow to your needs - Change SMS content: Edit the text in the “Send SMS reminder” node to match your team’s tone. - Modify call wait time: Enable and adjust the “Wait 5 minutes” node to any delay you prefer. - Add CRM integration: Replace or extend the Google Sheets node to update your CRM instead of a spreadsheet. - Customize call summary prompts: Edit the prompt sent to OpenAI to change the summary style or add extra insights. - Send email to different recipients: Change the recipient address in the Gmail node or make it dynamic from the lead record. Need help customizing? Contact me for consulting and support.

🌐 Confluence Page AI Chatbot Workflow This n8n workflow template enables users to interact with an AI-powered chatbot designed to retrieve, process, and analyze content from Confluence pages. By leveraging Confluence's REST API and an AI agent, the workflow facilitates seamless communication and contextual insights based on Confluence page data. 🌟 How the Workflow Works 🔗 Input Chat Message The workflow begins when a user sends a chat message containing a query or request for information about a specific Confluence page. 📄 Data Retrieval The workflow uses the Confluence REST API to fetch page details by ID, including its body in the desired format (e.g., storage, view). The retrieved HTML content is converted into Markdown for easier processing. 🤖 AI Agent Interaction An AI-powered agent processes the Markdown content and provides dynamic responses to user queries. The agent is context-aware, ensuring accurate and relevant answers based on the Confluence page's content. 💬 Dynamic Responses Users can interact with the chatbot to: - Summarize the page's content. - Extract specific details or sections. - Clarify complex information. - Analyze key points or insights. 🚀 Use Cases - Knowledge Management: Quickly access and analyze information stored in Confluence without manually searching through pages. - Team Collaboration: Facilitate discussions by summarizing or explaining page content during team chats. - Research and Documentation: Extract critical insights from large documentation repositories for efficient decision-making. - Accessibility: Provide an alternative way to interact with Confluence content for users who prefer conversational interfaces. 🛠️ Resources for Getting StartedConfluence API Setup: Generate an API token for authentication via Atlassian's account management portal. Refer to Confluence's REST API documentation for endpoint details and usage instructions. n8n Installation: Install n8n locally or on a server using the official installation guide. AI Agent Configuration: Set up OpenAI or other supported language models for natural language processing.

Who Is This For? This workflow is designed for AI engineers, automation specialists, and content creators who need a scalable system to dynamically manage prompts stored in GitHub. It eliminates manual updates, enforces required variable checks, and ensures that AI interactions always receive fully processed prompts. 🚀 What Problem Does This Solve? Manually managing AI prompts can be inefficient and error-prone. This workflow: ✅ Fetches dynamic prompts from GitHub ✅ Auto-populates placeholders with values from the setVars node ✅ Ensures all required variables are present before execution ✅ Processes the formatted prompt through an AI agent🛠 How This Workflow Works This workflow consists of three key branches, ensuring smooth prompt retrieval, variable validation, and AI processing. 1️⃣ Retrieve the Prompt from GitHub The workflow starts manually or via an external trigger. It fetches a text-based prompt stored in a GitHub repository. The Extract from File Node retrieves the content from the GitHub file. The SetPrompt Node stores the prompt, making it accessible for processing. 📌 Note: The prompt must contain n8n expression format variables (e.g., {{ $json.company }}) so they can be dynamically replaced. 2️⃣ Extract & Auto-Populate Variables A Code Node scans the prompt for placeholders in the n8n expression format ({{ $json.variableName }}). The workflow compares required variables against the setVars node: ✅ If all variables are present, it proceeds to variable replacement. ❌ If any variables are missing, the workflow stops and returns an error listing them. The Replace Variables Node replaces all placeholders with values from setVars. 📌 Example of a properly formatted GitHub prompt: Hello {{ $json.company }}, your product {{ $json.features }} launches on {{ $json.launch_date }}. This ensures seamless replacement when processed in n8n. 3️⃣ AI Processing & Output The Set Completed Prompt Node stores the final, processed prompt. The AI Agent Node (Ollama Chat Model) processes the prompt. The Prompt Output Node returns the fully formatted response. 📌 Optional: Modify this to use OpenAI, Claude, or other AI models. ⚠️ Error Handling: Missing Variables If a required variable is missing, the workflow stops execution and provides an error message: ⚠️ Missing Required Variables: ["launch_date"] This ensures no incomplete prompts are sent to AI agents. ✅ Example Use Case 📜 GitHub Prompt File (Using n8n Expressions) Hello {{ $json.company }}, your product {{ $json.features }} launches on {{ $json.launch_date }}. 🔹 Variables in setVars Node { "company": "PropTechPro", "features": "AI-powered Property Management", "launch_date": "March 15, 2025" } ✅ Successful Output Hello PropTechPro, your product AI-powered Property Management launches on March 15, 2025. 🚨 Error Output (If Missing launch_date) ⚠️ Missing Required Variables: ["launch_date"] 🔧 Setup Instructions1️⃣ Connect Your GitHub Repository Store your prompt in a public or private GitHub repo. The workflow will fetch the raw file using the GitHub API. 2️⃣ Configure the SetVars Node Define the required variables in the SetVars Node. Make sure the variable names match those used in the prompt. 3️⃣ Test & Run Click Test Workflow to execute. If variables are missing, it will show an error. If everything is correct, it will output the fully formatted prompt. ⚡ How to Customize This Workflow 💡 Need CRM or Database Integration? Connect the setVars node to an Airtable, Google Sheets, or HubSpot API to pull variables dynamically. 💡 Want to Modify the AI Model? Replace the Ollama Chat Model with OpenAI, Claude, or a custom LLM endpoint. 📌 Why Use This Workflow? ✅ No Manual Updates Required – Fetches prompts dynamically from GitHub. ✅ Prevents Broken Prompts – Ensures required variables exist before execution. ✅ Works for Any Use Case – Handles AI chat prompts, marketing messages, and chatbot scripts. ✅ Compatible with All n8n Deployments – Works on Cloud, Self-Hosted, and Desktop versions.

This flow is supported by a Chrome plugin created with Cursor AI. The idea was to create a Chrome plugin and a backend service in N8N to do chart analytics with OpenAI. It's a good sample on how to submit a screenshot from the browser to N8N. Who is it for? N8N developers who want to learn about using a Chrome plugin, an N8N webhook, and OpenAI. What opportunity does it present? This sample opens up a whole range of N8N connected Chrome extensions that can analyze screenshots by using OpenAI. What this workflow does? The workflow contains: - A webhook trigger - An OpenAI node with GPT-4O-MINI and Analyze Image selected - A response node to send back the text that was created after analyzing the screenshot. All this is needed to talk to the Chrome extension which is created with Cursor AI. The idea is to visit the tradingview.com crypto charts, click the Chrome plugin, and get back analytics about the shown chart in understandable language. This is driven by the N8N flow. With the new image analytics capabilities of OpenAI, this opens up a world of opportunities. Requirements/setup: - OpenAI API key - Cursor AI installed - The Chrome extension How to customize it to your needs? Both the Chrome extension and N8N flow can be adapted to use on other websites. You can consider: - Analyzing a financial screen and asking questions about the data shown - Analyzing other charts - Extending the N8N workflow with other AI nodes With AI and image analytics, the sky is the limit, and in some cases, it saves you from creating complex API integrations.

Who this is for? The Automate Etsy Data Mining with Bright Data Scrape & Google Gemini workflow is designed for eCommerce analysts, product researchers, and AI developers seeking to extract actionable insights from Etsy listings at scale. It is ideal for: - eCommerce Entrepreneurs - Researching product demand and competition. - Market Analysts - Tracking pricing, reviews, and trends across Etsy categories. - Product Managers - Identifying niche opportunities and design inspirations. - Data Scientists & AI Engineers - Automating product intelligence pipelines. - Growth Hackers - Leveraging Etsy insights to refine product-market fit. What problem is this workflow solving? Manually browsing Etsy to analyze product listings, pricing, reviews, and seller activity is slow, inconsistent, and unscalable. Scraping Etsy requires unlocking JavaScript-heavy content and structuring noisy data for analysis. This workflow solves: - Automated and scalable scraping of Etsy product listings using Bright Data’s infrastructure. - A fully paginated data structured Etsy production data extraction via the Google Gemini LLM. - Enables faster decision-making for product research and competitive analysis via the fully automated paginated data extraction. What this workflow does - Receives input: Sets the Etsy URL for the data extraction and analysis. - Uses Bright Data's Web Unlocker to extract content from relevant sites. - Cleans and preprocesses the scraped content for readability. - Sends the content to Google Gemini for enriched results including data persistence over the disk. - Sends the response to a target system via Webhook notification. Setup 1. Sign up at Bright Data. 2. Navigate to Proxies & Scraping and create a new Web Unlocker zone by selecting Web Unlocker API under Scraping Solutions. 3. In n8n, configure the Header Auth account under Credentials (Generic Auth Type: Header Authentication). The Value field should be set with the Bearer XXXXXXXXXXXXXX. The XXXXXXXXXXXXXX should be replaced by the Web Unlocker Token. 4. A Google Gemini API key (or access through Vertex AI or proxy). 5. Update the Set Etsy Search Query for setting the brand content URL and the Bright Data Zone name. 6. Update the Webhook HTTP Request node with the Webhook endpoint of your choice. How to customize this workflow to your needs - Input Sources: Replace the static URL with dynamic input from Google Sheets, Webhook, or Airtable to research multiple niches. - Prompt Customization: Adjust Gemini prompts to extract specific insights, for example: - List key features of the product - Summarization of the review themes - Data Output Options: Update the Webhook notification to save data to: - Google Sheets - Notion or Airtable - SQL/NoSQL - Slack/Email

Create a form for customer feedback, have OpenAI classify the sentiment of the feedback (positive/neutral/negative) and store it in Google Sheets! Set up steps: 1. Connect Google Sheets 2. Connect your OpenAI account (API key + Org ID) 3. Create a customer feedback form, use an existing one or use the one below as an example. All set! Here is the example Google Sheet being used in this workflow. You can download it to your account.

Enhance your data analysis by connecting an AI Agent to your dataset using n8n tools. This template teaches you how to build an AI Data Analyst Chatbot that is capable of pulling data from your sources, using tools like Google Sheets or databases. It's designed to be easy and efficient, making it a good starting point for AI-driven data analysis. You can easily replace the current Google Sheets tools for databases like Postgres or MySQL. How It Works The core of the workflow is the AI Agent. It's connected to different data retrieval tools to get data from Google Sheets (or your preferred database) in many different ways. Once the data is retrieved, the Calculator tool allows the AI to perform mathematical operations, making your data analysis precise. Who is this template for - Data Analysts & Researchers: Pull data from different sources and perform quick calculations. - Developers & AI Enthusiasts: Learn to build your first AI Agent with easy dataset access. - Business Owners: Streamline your data analysis with AI insights and automate repetitive tasks. - Automation Experts: Enhance your automation skills by integrating AI with your existing databases. How to Set Up You can find detailed instructions in the workflow itself.

Automated Research Report Generation with OpenAI, Wikipedia, Google Search, Gmail/Telegram and PDF OutputWhat Problem Does This Solve? 🛠️ This workflow automates the process of generating professional research reports for researchers, students, and professionals. It eliminates manual research and report formatting by aggregating data, generating content with AI, and delivering the report as a PDF via Gmail or Telegram. Target audience: Researchers, students, educators, and professionals needing quick, formatted research reports. What Does It Do? 🌟 - Aggregates research data from Wikipedia, Google Search, and SerpApi. - Refines user queries and generates structured content using OpenAI. - Converts the content into a professional HTML report, then to PDF. - Sends the PDF report via Gmail or Telegram. Key Features 📋 - Real-time data aggregation from multiple sources. - AI-driven content generation with OpenAI. - Automated HTML-to-PDF conversion for professional reports. - Flexible delivery via Gmail or Telegram. - Error handling for robust execution.

This workflow is a highly advanced multimodal AI assistant designed to operate through WhatsApp. It can understand and respond to text, images, voice messages, and PDF documents by combining OpenAI models with smart logic to adapt to the content received. 🎯 Core Features📥 1. Automatic Message Type Detection Using the Input type node, the bot detects whether the user has sent: - Text - Voice messages - Images - Files (PDF) - Other unsupported content💬 2. Smart Text Message Handling Text messages are processed by an OpenAI GPT-4o-mini agent with a customized system prompt. Replies are concise, accurate, and formatted for mobile readability. 🖼️ 3. Image Analysis & Description Images are downloaded, converted to base64, and analyzed by an image-aware AI model. The output is a rich, structured description, designed for visually impaired users or visual content interpretation. 🎙️ 4. Voice Message Transcription & Reply Audio messages are downloaded and transcribed using OpenAI Whisper. The transcribed text is analyzed and answered by the AI. Optionally, the AI reply can be converted back to voice using OpenAI's text-to-speech and sent as an audio message. 📄 5. PDF Document Extraction & Summary Only PDFs are allowed (filtered via MIME type). The document’s content is extracted and combined with the user's message. The AI then provides a relevant summary or answer. 🧠 6. Contextual Memory Each user has a personalized session ID with a memory window of 10 interactions. This ensures a more natural and contextual conversation flow. How It Works This workflow is designed to handle incoming WhatsApp messages and process different types of inputs (text, audio, images, and PDF documents) using AI-powered analysis. Here’s how it functions: - Trigger: The workflow starts with the WhatsApp Trigger node, which listens for incoming messages (text, audio, images, or documents). - Input Routing: The Input type (Switch node) checks the message type and routes it to the appropriate processing branch: - Text: Directly forwards the message to the AI agent for response generation. - Audio: Downloads the audio file, transcribes it using OpenAI, and sends the transcription to the AI agent. - Image: Downloads the image, analyzes it with OpenAI’s GPT-4 model, and generates a detailed description. - PDF Document: Downloads the file, extracts text, and processes it with the AI agent. - Unsupported Formats: Sends an error message if the input is not supported. - AI Processing: The AI Agent1 node, powered by OpenAI, processes the input (text, transcribed audio, image description, or PDF content) and generates a response. - Response Handling: - For audio inputs, the AI’s response is converted back into speech (using OpenAI’s TTS) and sent as a voice message. - For other inputs, the response is sent as a text message via WhatsApp. - Memory: The Simple Memory node maintains conversation context for follow-up interactions. Setup Steps To deploy this workflow in n8n, follow these steps: 1. Configure WhatsApp API Credentials: Set up WhatsApp Business API credentials (Meta Developer Account). Add the credentials in the WhatsApp Trigger, Get Image/Audio/File URL, and Send Message nodes. 2. Set Up OpenAI Integration: Provide an OpenAI API key in the Analyze Image, Transcribe Audio, Generate Audio Response, and AI Agent1 nodes. 3. Adjust Input Handling (Optional): Modify the Switch node (“Input type”) to handle additional message types if needed. Update the “Only PDF File” IF node to support other document formats. 4. Test & Deploy: Activate the workflow and test with different message types (text, audio, image, PDF). Ensure responses are correctly generated and sent back via WhatsApp. Need help customizing? Contact me for consulting and support or add me on Linkedin.

This workflow shows how to use a self-hosted Large Language Model (LLM) with n8n's LangChain integration to extract personal information from user input. This is particularly useful for enterprise environments where data privacy is crucial, as it allows sensitive information to be processed locally. 📖 For a detailed explanation and more insights on using open-source LLMs with n8n, take a look at our comprehensive guide on open-source LLMs. Key Features - Local LLM Connect Ollama to run Mistral NeMo LLM locally. Provide a foundation for compliant data processing, keeping sensitive information on-premises. - Data extraction Convert unstructured text to a consistent JSON format. Adjust the JSON schema to meet your specific data extraction needs. - Error handling Implement auto-fixing for LLM outputs. Include error output for further processing. Setup and ConfigurationPrerequisites n8n AI Starter Kit installed. Configuration steps 1. Add the Basic LLM Chain node with system prompts. 2. Set up the Ollama Chat Model with optimized parameters. 3. Define the JSON schema in the Structured Output Parser node. 🔍 Further resources Run LLMs locally with n8n. Video tutorial on using local AI with n8n. Apply the power of self-hosted LLMs in your n8n workflows while maintaining control over your data processing pipeline!

The n8n Nostr Community Node is a tool that integrates Nostr functionality into n8n workflows, allowing users to interact with the Nostr protocol seamlessly. It provides both read and write capabilities and can be used for various automation tasks. This node is ideal for self-hosted n8n setups, as community nodes are not supported on n8n cloud. It opens up exciting possibilities for integrating workflows with the decentralized Nostr protocol. ### Features - Write Operations: Send notes and events (kind1) to the Nostr network. - Read Operations: Fetch events based on criteria such as event ID, public key, hashtags, mentions, or search terms. - Utility Functions: Convert events into different formats like naddr or nevent and handle key transformations between bech32 and hex formats. - Trigger Events: Monitor the Nostr network for specific mentions or events and trigger workflows automatically. ### Use Cases - Automating note posting without exposing private keys. - Setting up notifications for mentions or specific events. - Creating bots or AI assistants that respond to mentions on Nostr. ### Installation 1. Install n8n on your system. 2. Add the Nostr Community Node to your instance. 3. Configure your credentials using a Nostr secret key (supports bech32 or hex formats).

This template updates Google Sheets rows with AI-generated responses and parsed JSON from every new row of data on your sheet.