Workflow Templates

How it works: Automatically detects when a new receipt is uploaded to Google Drive. Extracts text from the receipt using OCR. Uses an AI Agent to analyze the extracted data and structure it (e.g., vendor, date, total, tax). Saves the organized receipt data into a Google Sheet for easy tracking. Set up steps: Setup takes around 15–20 minutes. You'll need a Google Drive folder for receipts and a Google Sheet to store results. Configure your Google Drive Trigger, OCR extraction, AI Agent, and Google Sheets connection. Detailed instructions and explanations are included in this n8n Starter Session tutorial series.

How it works This workflow automatically analyzes meeting effectiveness and provides constructive feedback: • Retrieve meeting minutes: Automatically searches and retrieves meeting minutes from Google Drive using either a Google Docs URL or meeting name • Multi-dimensional analysis: Comprehensively evaluates meeting effectiveness score, speaking time distribution, communication quality (clarity, friendliness, decisiveness, listening), disagreements, and more • Generate actionable feedback: Outputs a structured report in Japanese with specific improvement suggestions and highlights of what went well Set up steps Setup takes approximately 5 minutes: • Connect Google Drive: Grant permission for the workflow to access your meeting minutes by connecting to Google Drive • First run: Enter the meeting minutes URL or meeting name and execute the workflow For detailed setup instructions and step-by-step explanations, please refer to the sticky notes inside your workflow.

Automatically Classify Support Tickets in Zoho Desk with AI with Gemini Transform your customer support workflow with intelligent ticket classification. This automation leverages AI to automatically categorize incoming support tickets in Zoho Desk, reducing manual work and ensuring faster ticket routing to the right teams. How It Works Fetches all tickets from Zoho Desk with pagination support Filters unclassified tickets (where classification field is null) Retrieves complete ticket threads for full conversation context Uses OpenRouter AI (GPT-4, Claude, or other models) to classify tickets into predefined categories Updates tickets in Zoho Desk with accurate classifications automatically Use Cases Customer Support Teams**: Automatically route tickets to specialized departments (billing, technical, sales) Help Desks**: Prioritize urgent issues and categorize feature requests Prerequisites Active Zoho Desk account with API access OpenRouter API account (supports multiple AI models) Basic understanding of OAuth2 authentication Predefined ticket categories in your Zoho Desk setup Setup Steps Time: ~15 minutes Configure Zoho Desk OAuth2 - Follow our step-by-step GitHub guide for OAuth2 credential setup Set up OpenRouter API - Create an account and generate API keys at openrouter.ai Customize classifications - Define your ticket categories (e.g., Technical, Billing, Feature Request, Bug Report) Adapt the workflow - Modify for any field: status, priority, tags, assignment, or custom fields Review API documentation - Check Zoho Desk Search API docs for advanced filtering options Test thoroughly - Run manual triggers before automation Note: This workflow demonstrates proper Zoho Desk API integration, including OAuth2 authentication and pagination handling—two common integration challenges.

What This Does Automatically finds relevant Reddit posts where your brand can add value, generates helpful AI comments, and sends the best opportunities to your Slack channel for review. Setup Requirements Reddit API credentials OpenAI API key Slack webhook URL Quick Setup Reddit API Create app at reddit.com/prefs/apps (select "script" type) Add client ID and secret to n8n credentials Configure Subreddits Edit the workflow to monitor subreddits relevant to your business: entrepreneur, startups, smallbusiness, [your_niche] AI Prompt Setup Customize the OpenAI node with your brand context: You're helping in [subreddit] discussions. When relevant, mention how [your_product] solves similar problems. Be helpful first, promotional second. Slack Integration Add your webhook URL to get notifications with: Post title and link AI-generated comment Engagement score (1-10) Key Features Smart Filtering**: AI evaluates if a post is worth engaging with Brand-Aware Comments**: Generated responses stay on-brand and helpful Team Review**: All opportunities go to Slack before posting Multiple Subreddits**: Monitor several communities simultaneously Customization Tips Adjust AI Scoring - Modify what makes a "good" opportunity: Post engagement level Relevance to your product Tone of the discussion Comment Templates - Set different styles for different subreddits: Technical advice for developer communities Business insights for entrepreneur groups User experience for product discussions Best Practices Start with 2-3 subreddits to test effectiveness Review and approve comments in Slack before posting Follow Reddit's 90/10 rule (90% helpful content, 10% self-promotion) Adjust the AI prompt based on what works in your communities Why Use This Saves hours of manual Reddit browsing Maintains consistent brand voice Never miss relevant conversations Team can review before engaging publicly

🛍️ Automated Product Description Generation for Adobe Commerce (Magento 2) Description This n8n template demonstrates how to automatically generate product descriptions for items in Adobe Commerce (Magento 2) that are missing one. The workflow retrieves product data, converts raw attribute values (like numeric IDs) into human-readable labels, and passes the enriched product data to an LLM (Azure OpenAI by default). The LLM generates a compelling description, which is then saved back to Magento using the API. This ensures all products have professional descriptions without manual writing effort. Use cases include: Auto-generating missing descriptions for catalog completeness. Creating consistent descriptions across large product datasets. Reducing manual workload for content managers. Tailoring descriptions for SEO and customer readability. Good to know All attribute options are resolved to human-readable labels before being sent to the LLM. The flow uses Azure OpenAI, but you can replace it with OpenAI, Anthropic, Gemini, or other LLM providers. The LLM prompt can be customised to adjust tone, length, SEO-focus, or specific brand style. Works out-of-the-box with Adobe Commerce (Magento 2) APIs, but can be adapted for other ecommerce systems. How it works Get Product from Magento Retrieves a product that has no description. Collects all product attributes. Generate Description with LLM Resolves attribute option IDs into human-readable values (e.g. color_id = 23 → "Red"). Passes the readable product attributes to an Azure OpenAI model. The LLM creates a clear, engaging product description. The prompt can be customised (e.g. SEO-optimized, short catalog text, or marketing style). Save Description in Magento Updates the product via the Magento API with the generated description. Ensures product data is enriched and visible in the webshop immediately. How to use Configure your Magento 2 API credentials in n8n. Replace the Azure OpenAI node with another provider if needed. Adjust the prompt to match your brand’s tone of voice. Run the workflow to automatically process products missing descriptions. Requirements ✅ n8n instance (self-hosted or cloud) ✅ Adobe Commerce (Magento 2) instance with API access ✅ Azure OpenAI (or other LLM provider) credentials (Optional) Prompt customisations for SEO or brand voice Customising this workflow This workflow can be adapted for: Other attributes**: Include or exclude attributes (e.g. only color & size for apparel). Different LLMs**: Swap Azure OpenAI for OpenAI, Anthropic, Gemini, or any supported n8n AI node. Prompt tuning**: Adjust instructions to generate shorter, longer, or SEO-rich descriptions. Selective updates**: Target only specific categories (e.g. electronics, fashion). Multi-language support**: Generate product descriptions in multiple languages for international shops.

This workflow automatically analyses tasks to uncover why the actual time spent exceeds the original estimates. It connects with ClickUp(Can do with any PMS like JIRA, Asana, Monday and more) and other project management tools to generate clear insights on overspending trends. Save time, improve planning accuracy, and boost team productivity with automated task time analysis with two types of reports. “Why needed extra time?”** – Reasons users requested extensions or faced blockers. “Why went over estimate?”** – Reasons the actual work exceeded the original estimation. 🔧 Workflow Overview Manual Trigger Kick off execution by clicking “Test workflow”. Fetch Relevant Tasks Calls ClickUp to retrieve all tasks in specified states (“internal review” or “in progress”) that belong to designated folders and assignees. Filter by Overrun Filters tasks to include only those where time_spent > time_estimate. Gather Details For each overrun task: Fetch time entries via ClickUp API. Fetch all comments, including threaded replies. Retain only essential task fields and reformat timestamps. Normalize and Merge Extracts and sorts comment threads into clean arrays. Sorts time entry intervals chronologically. Merges task metadata, comments, and time entries into a single payload. Pass to AI Agent Sends consolidated task data to a ChatGPT-powered node using a custom prompt that: Extracts all “extra time requests” from comments and time entries. Identifies debugging, research, clarification, or rework intervals exceeding estimates. Combines findings into two distinct checklists. Format JSON Output A final Code node parses AI output into a clean JSON array ready for conversion. Convert to File JSON result for each task is prepared for file attachment or external storage. 🧩 Key Nodes & Functions |Node Name|Responsibility| |-|-| |Get ClickUp Tasks|Retrieves tasks by filter criteria| |If task has crossed estimation|Ensures only tasks with overruns continue| |Fetch Time entries via task IDs|Retrieves detailed time intervals| |Fetch Master comments|Retrieves all comments and threads| |Split → Merge scripting nodes|Clean and normalize comments structure| |Modify Time/Task data|Trims and prepares JSON for AI processing| |OpenAI Chat Model + AI Agent|Applies a GPT-based prompt to generate two reasoned checklists in JSON format| |Convert to File|Prepares final output as a JSON file or store on Sreadsheet or Email or Excel| 🛠 Customization Tips Trigger Automation:** Integrate a scheduled node for periodic runs (e.g., daily). Filter Scope:** Adjust ClickUp filters for different task types, spaces, or statuses. AI Prompt Tuning:** Refine prompt to include severity, link references, or categorize reasons. Output Handling:* Use the JSON file in subsequent n8n nodes for notifications *(Slack, Email, Spreadsheet, Airtable, ExcelSheet, etc.) or analytics. ✅ Benefits at a Glance Automates time-overrun analysis, eliminating manual review. Extracts insights directly from tasks description, comments, and timesheets. Produces structured outputs ideal for management dashboards or retrospectives. Customizable for team-specific workflows or reporting needs. 🔐 API Credentials Needed You will required to create API key of your ClickUp Account. Follow the n8n instruction document here ++https://docs.n8n.io/integrations/builtin/credentials/clickup/++ this will guide you how you can connect your ClickUp acount with n8n workflow. 👨🏻‍💻 Need Help? Contact www.KrupalPatel.com for support and custom workflow development. Find more n8n workflow for real world use cases from here: ++https://n8n.io/creators/krupalpatel/++

AI Interior Design Assistant: Your Digital Design Partner What This System Does This n8n workflow transforms your Telegram into a professional interior design studio powered by artificial intelligence. Send a photo of furniture or a room space, and watch as the system intelligently catalogs items, documents spaces, and generates stunning custom interior designs tailored to your vision. The Magic Behind the Scenes Smart Image Recognition When you upload a photo, the system immediately springs into action with sophisticated image analysis: Furniture Detection**: Spots individual pieces like sofas, chairs, tables, and lamps with catalog-precision accuracy Room Analysis**: Identifies complete spaces, architectural features, lighting conditions, and existing design elements Style Classification**: Determines design styles from modern minimalist to traditional classic Material Recognition**: Identifies wood types, fabric textures, metal finishes, and color palettes Intelligent Database Management The workflow maintains three interconnected databases that work like a professional design firm's catalog system: Furniture Catalog (catalog_products) Comprehensive product details including style, materials, dimensions, and compatibility Professional descriptions written for interior designers Searchable tags for quick design matching High-quality image storage for visual reference Room Documentation (rooms) Detailed space analysis including size, style, and architectural features Color palette documentation and lighting assessment Existing furniture inventory for design planning Room-specific design recommendations Design Portfolio (ai_generated_images) Archive of all AI-generated interior designs Original prompts and design descriptions Searchable by style, room type, or specific elements Ready for client presentations or further modifications AI-Powered Design Generation The system's crown jewel is its ability to create stunning interior visualizations: Contextual Understanding: Combines room characteristics with catalog products to create realistic design scenarios Professional Prompting: Generates detailed, interior-design-specific prompts that result in high-quality, commercially viable designs Style Consistency: Maintains design coherence across different elements while respecting user preferences Modification Capabilities: Can reference and modify previous designs, allowing for iterative improvements The User Experience Journey Scenario 1: Building Your Furniture Catalog Upload: Send photos of furniture pieces via Telegram Analysis: AI examines each piece, identifying style, materials, dimensions, and design era Cataloging: Items are professionally documented with searchable metadata Confirmation: Receive detailed catalog entries for each piece Scenario 2: Documenting Your Spaces Room Photos: Share images of your living spaces Space Analysis: AI assesses room size, style, lighting, and architectural features Documentation: Complete room profiles are created for design planning Inventory: Existing furniture and design elements are noted Scenario 3: Creating Custom Designs Design Request: Ask for specific interior modifications or new layouts Smart Matching: System pulls relevant items from your catalog and room data AI Generation: Gemini 2.5 Flash creates photorealistic interior designs Instant Delivery: Receive professional-quality visualizations via Telegram Scenario 4: Design Evolution Reference Previous Work: Mention earlier designs you want to modify Contextual Modification: AI understands your reference and applies new changes Enhanced Generation: Creates updated designs building on previous concepts Continuous Improvement: Iterate until the design matches your vision Technical Sophistication Multi-AI Coordination OpenAI GPT-4**: Handles complex reasoning, database operations, and user interaction Google Gemini 2.5 Flash**: Specializes in high-quality image generation with interior design expertise Intelligent Routing**: Automatically determines whether to catalog, document, or generate based on context Professional Data Structure The database schema reflects real interior design workflows: Industry-standard categorization systems Professional terminology and measurements Design compatibility matrices Style and era classifications used by actual designers Seamless Integration Telegram Interface**: No app downloads or complex interfaces - just send photos and text Cloud Storage**: All images stored in Supabase with public URLs for easy access Real-time Processing**: Immediate feedback and rapid design generation Persistent Memory**: Everything is saved and searchable for future reference Why This Matters This workflow bridges the gap between professional interior design tools and accessible consumer technology. It provides: For Design Professionals: A powerful cataloging and visualization tool that streamlines client presentations and design iteration For Homeowners: Professional-level design capability without the cost or complexity of traditional design software For Businesses: A scalable solution for furniture visualization, space planning, and customer engagement The Innovation Factor Unlike simple design apps that work with generic templates, this system: Learns your specific furniture and spaces Maintains design continuity across projects Provides professional-quality outputs Scales from single rooms to complete home designs Integrates seamlessly into your daily communication workflow The result is a design assistant that feels less like software and more like having a professional interior designer available 24/7 through your phone. Future Possibilities This foundation supports expansion into: Room dimension calculations and space optimization Integration with furniture retailers for purchase links 3D room modeling and virtual reality previews Style preference learning and automated suggestions Multi-user collaboration for design teams The workflow represents a new paradigm where AI doesn't replace human creativity but amplifies it, making professional design capabilities accessible to anyone with a smartphone and an imagination.

Overview Turn your docs into an AI-powered internal or public-facing assistant. This chatbot workflow uses RAG (Retrieval-Augmented Generation) with Supabase vector search to answer employee or customer questions based on your company documents—automatically updated via Google Drive. Whether it’s deployed in Telegram or embedded on your website, this agent supports voice and text input, transcribes voice messages, pulls relevant context from your internal files, and responds with a helpful, AI-generated answer. Two additional workflows listen for file changes in a shared Google Drive folder, convert them into embeddings using OpenAI, and sync them with your Supabase vector DB—so your knowledge base is always up to date. Who’s it for Startups building an internal ops or HR assistant SaaS companies deploying help bots on their websites Customer support teams reducing repetitive questions Knowledge-driven teams needing internal AI assistants How it works Triggered via Telegram bot (or easily swapped for website chatbot or “on chat message”) If user sends a voice message, it’s transcribed to text using OpenAI Whisper Input is passed to a RAG agent that: Searches a Supabase vector store for relevant docs Pulls context from matching chunks using OpenAI embeddings Responds with an LLM-powered answer The response is sent back as a Telegram message Two separate workflows: New File Workflow: Listens for file uploads in Google Drive, extracts and splits text, then sends to Supabase with embeddings Update File Workflow: Detects file edits, deletes old rows, and updates embeddings for the revised file Example use case > You upload your internal policy docs and client FAQs into a Google Drive folder. > > Employees or customers can now ask: > - “What’s the refund policy for annual plans?” > - “How do I request a day off?” > - “What tools are approved for use by the engineering team?” > > The chatbot instantly pulls up the right section and responds with a smart, confident answer. How to set up Connect a Telegram bot or use n8n’s webchat / chatbot widget Hook up OpenAI for transcription, embeddings, and completion Set up a Supabase project and connect it as a vector store Upload your internal docs to Google Drive Deploy the “Add File” and “Update File” automations to manage embedding sync Customize the chatbot’s tone and personality with prompt tweaks Requirements Telegram bot (or n8n Chat widget) Google Drive integration Supabase with pgvector or similar enabled OpenAI API key (Whisper, Embeddings, ChatGPT) Two folders: one for raw documents and one for tracking updates How to customize Swap Supabase for Pinecone, Weaviate, or Qdrant Replace Telegram with web chat, Slack, Intercom, or Discord Add logic to handle fallback answers or escalate to human Embed the chat widget on your site for public customer use Add filters (e.g. department, date, author) to narrow down context

Overview This n8n workflow automates the entire process of capturing leads, enriching their data with company information using an AI Agent, and then generating highly personalized introductory emails (using ChatGPT-4o) saved as drafts in your Gmail account. This prepares your sales team for a high-quality outreach with minimal manual effort. Requirements To use this workflow, you need the following accounts and credentials: Google Sheets Account: To store and track lead information (the workflow uses a sheet with ID). Below are the columns of the sheet First name Last name Email ID Company Name Company Information Designation Message Location Status Intro email Date Reminder 1 needed? Reminder 1 Email Date OpenAI API Key (for ChatGPT-4o): For drafting the personalized introductory emails. Google Gemini API Key: For the AI Agent to perform online company research. Gmail Account: To save the final personalized emails as drafts. How It Works The workflow is structured into two main phases: Lead Capture & Enrichment, and Personalized Email Drafting. Phase 1: Lead Capture and Enrichment This phase collects user inquiries and uses an AI Agent to search the web for additional company details to enrich the lead profile. On form submission (Form Trigger): The workflow starts when a potential lead fills out the embedded lead capture form, which collects details like First Name, Last Name, Company Name, Email ID, Designation, and a Message/inquiry. This is optional as many company may have other ways to capture leads. Append row in sheet (Google Sheets): The initial lead data collected from the form is added to your Google Sheet tracker, setting the Status to To Send. AI Agent: The AI Agent is prompted to search online for the client's company name to gather two pieces of information: A 1-2 sentence Company Description (what they do). The Company Location, categorized as Delhi/NCR, Bangalore, Mumbai, or Other. This should be changed basis your need. Code: This node processes the structured text output from the AI Agent and separates the Company Description and Company Location into distinct fields. Update row in sheet (Google Sheets): The newly researched Company Information and Location are updated and added to the lead's row in the Google Sheet, matching on Email ID. Phase 2: Personalized Email Drafting and Logging This phase retrieves leads ready for outreach, drafts a personalized email using AI, and saves it for the sales team. Get row(s) in sheet (Google Sheets): The workflow fetches all leads whose Status is either To send or To Send (using an OR filter). Introductory email (OpenAI - ChatGPT-4o): For each lead, the OpenAI node is used as a B2B marketing assistant to write a personalized introductory email based on a predefined template. The prompt uses the lead's data (First Name, Company Name, Message, etc.) and instructs the AI to: Create a subject line: Following up on your interest in <your company name> for [shorter version of pain point]. Personalize the body by referencing their pain points and suggesting how <your company> has helped similar companies. Include a call-to-action (CTA) for a quick 15-minute chat. Provide a P.S. line about a relevant success story that your company has delivered. The output is structured into EmailSubject, EmailContent, and Emailid variables. Create a draft (Gmail): The personalized email is saved as a draft in the specified Gmail account, using the AI-generated Subject and Content. Best Practice: It is recommended to add an auto-signature in the Gmail account used for the draft. Append or update row in sheet (Google Sheets): The lead's row is updated to reflect the outreach effort. The Status is set to Drafted, and the current date is logged in the Intro email Date column. Customization Notes Initial Data: You can replace the On form submission trigger with a Google Sheets Trigger or a Webhook to capture leads from other sources (e.g., a CRM or LinkedIn). AI Prompt: To ensure the best results, update the agent prompt in the Introductory email node to make it more relevant for your company. Sender: Ensure the email ID used for drafting corresponds to the sales team's email.

Build Your First AI Agent with ChatGPT-5 Who is this for? This workflow is designed for beginners and professionals who want to build their first AI-powered assistant with n8n. It’s perfect for anyone managing online trainings, consultations, or services that require both a knowledge base and appointment scheduling. What problem is this workflow solving? Manually handling client questions, checking your availability, and confirming bookings can be time-consuming and error-prone. This workflow automates the process, ensuring quick, accurate answers and seamless scheduling directly through chat. What this workflow does Answers user questions using your knowledge base stored in Google Sheets. Checks availability in Google Calendar and proposes alternative time slots if needed. Books 1-hour appointments in Paris time only after client confirmation. Sends a professional confirmation email with all appointment details. Setup Import this workflow into your n8n instance. Connect your Google Sheets, Gmail, and Google Calendar credentials. Add your knowledge base into Google Sheets (questions, answers, policies, packs, etc.). Test the workflow using the Connected Chat Trigger node to start conversations with the AI Agent. How to customize this workflow to your needs Update the Google Sheets database with your own training packs, services, or company FAQs. Adjust the email template to reflect your branding and communication style. Modify the appointment duration if you need sessions longer or shorter than 1 hour. Add extra nodes (e.g., CRM integration) to capture leads or sync appointments with external systems. 📄 Documentation: Notion Guide Need help customizing? Contact me for consulting and support : Linkedin / Youtube

AI-Powered Keyword Cannibalization Detection Workflow Overview This is an advanced n8n automation workflow designed to detect and analyze keyword cannibalization issues across multiple client websites using Google Search Console data and artificial intelligence. The system provides real-time monitoring and comprehensive reporting to help SEO professionals identify and resolve internal competition between pages ranking for the same keywords. Core Components 1. Automated Monitoring System Real-time trigger:** Monitors Google Sheets for keyword changes every minute Multi-client support:** Handles up to 4 different client websites simultaneously Intelligent routing:** Automatically directs each client's data through dedicated processing paths 2. Data Collection & Processing GSC Integration:** Fetches 30 days of search performance data from Google Search Console API Comprehensive metrics:** Collects keyword rankings, page URLs, positions, clicks, impressions, and CTR Data transformation:** Groups raw API responses by keywords for structured analysis Cross-referencing:** Matches target keywords from Google Sheets with actual GSC performance data 3. AI Analysis Engine GPT-4o powered:** Uses advanced AI to analyze keyword competition patterns Risk categorization:** Automatically classifies cannibalization risk as: High Risk: 5+ pages competing for the same keyword Moderate Risk: 3+ pages ranking in top 10 positions Low Risk: 2 pages with one clearly dominating No Risk: Single page ranking for the keyword Intelligent reasoning:** Provides detailed explanations for each risk assessment 4. Comprehensive Reporting Automated output:** Saves analysis results back to Google Sheets Detailed insights:** Includes risk levels, reasoning, observations, and actionable remediation steps Performance tracking:** Complete keyword performance metrics for client reporting Status tracking:** Identifies which keywords are ranking vs. missing from search results

If you want to reach a wider audience, having your video titles and descriptions in multiple languages can help you connect with more viewers. This template provides the configuration needed to generate translations and update them directly on YouTube. How it works Defines Video ID and Languages** You can get the ID from the video URL. If you’re unsure, click the Share button, it will provide the ID at between the "/" and the "?si=" Fetches video information** As long as the workflow has the video ID, it can retrieve the video information, whether the video is published or not. Check languages to translate** If no default language is set in the video details, the workflow will assume "en" (English) as the default. To overwrite the default language, replace 'en' in line 2 of the code. AI Agent Translator**: You can improve results by refining the prompt. Feel free to experiment, just don’t change the formatting structure. Updates Video Localization** The API requires to send current default version again along with the translations in the Body. Output URL** returns the URL for the video’s localization. It may take a few seconds to appear—refresh and try again if needed. ⚠ The translation will overwrite any existing translations for the same language. How to Use (First time only) Set up credentials for your user. Edit the node "Defines Video ID and Languages". Click the Execute Workflow button. Enjoy your translated video titles and descriptions! More details can be found in the sticky notes under each node. Requirements To implement this workflow, you will need to configure credentials for the following nodes: Fetch Video Information**: YouTube Auth This of course, assumes that you have a YouTube account. The account creation and setup takes time, but luckily it's the only slow part, once done, you are good to go. AI Agent Translator**: Google Gemini (PaLM) API Both credentials can be created directly inside the node by clicking it, then selecting “Create Credential to Connect With” and following the instructions provided by the n8n Assistant. If you already have the credentials, simply select them! No paid plan is required, free tiers are sufficient. Keep in mind the request limits if you extend or modify this workflow to do multiple videos at once.