Workflow Templates

Tagged with: , , , ,

Chat with a Slack channel using AI. This workflow fetches the channel’s message history and lets you ask natural-language questions (“what were the decisions?”, “who’s blocked?”, “summarize yesterday”). The assistant only answers from the channel’s actual messages—no guessing. ⚙️ Setup Instructions 1️⃣ Set Up OpenAI Connection Go to OpenAI Platform Navigate to OpenAI Billing Add funds to your billing account Copy your API key into the OpenAI credentials in n8n 2️⃣ Connect Slack API Create an app → <https://api.slack.com/apps> OAuth & Permissions → add scopes you need to read channel history (typical: channels:history, groups:history, im:history, mpim:history, plus channels:read, groups:read, users:read. Add chat:write if you want the bot to reply in Slack.) Install the app to your workspace → copy the Bot User OAuth Token In n8n → Credentials → New → Slack OAuth2 API → paste token and save In the Slack History node, select your Slack credential and the Channel ID to read 🗣️ Example Questions You Can Ask “Give me a 5-bullet summary of the last 24 hours.” “What action items were assigned, and to whom?” “List open questions that haven’t been answered yet.” “Who was mentioned most this week?” “Summarize decisions from the last sprint planning.” “Show messages with the word ‘blocker’ from the past 2 days.” “What files/links were shared today?” 📬 Contact Need help customizing this or adding auto-replies back into Slack? 📧 robert@ynteractive.com 🔗 Robert Breen 🌐 ynteractive.com

Video Introduction Want to automate your inbox or need a custom workflow? 📞 Book a Call | 💬 DM me on Linkedin What This Workflow Does This workflow automatically enriches your Airtable company database with AI-powered research. Whenever a new company is added to your Airtable base, an AI agent searches the web to discover the company's website and generates a concise 1-2 sentence summary of their business activities. The workflow then updates the Airtable record with these findings, eliminating manual research and keeping your data complete and current. Key Benefits Automated data enrichment**: No more manual web searches for company information AI-powered accuracy**: Uses OpenAI with web search capabilities to find reliable, up-to-date information Structured output**: Returns data in a predictable JSON format for consistent record updates Real-time processing**: Triggers immediately when new companies are added Setup Requirements Airtable Configuration: Your Companies table must have a Created time field to trigger the workflow Create an Airtable Personal Access Token at airtable.com/create/tokens with these scopes: data.records:read, data.records:write, and schema.bases:read Ensure your table has fields for Website and Company Details (or modify the field mappings accordingly) OpenAI Setup: Register for an OpenAI account and obtain an API key Add your OpenAI credentials in n8n Workflow Configuration: Update the Airtable Trigger node with your base and table URLs Update the Airtable Update node with your specific base and table information Customize the AI prompt if you need different research parameters or output fields Once published, the workflow runs automatically whenever a new company is added to your table.

This n8n template automates the identification & optimisation of Answer Engine Optimisation (AEO) opportunities by extracting Google "People Also Ask" (PAA) data to generate high-authority response snippets. As search engines shift toward AI-generated overviews and answer engines like Perplexity, appearing in the "Citation" or "Snippet" layer is critical for maintaining organic visibility. This workflow identifies the exact questions users are asking and drafts structured answers designed to win featured snippets. How it works The workflow initiates via a Form Submission with a target keyword. It uses the SerpApi node to scrape real-time Google Search results, specifically targeting the related_questions (PAA) array. An Item Lists node cleans the data to ensure only high-relevance questions are processed. The questions are sent to a Gemini node with a specialized prompt to draft a 40-50 word answer optimized for "Speakable" Schema and featured snippet logic. The finalised Q&A pairs are then appended to a Google Sheet, creating a ready-to-publish content gap report for the agency's editorial team. How to use Replace the keyword placeholder in the HTTP/SerpApi node with your client’s target search term. Connect your Notion or Google Sheets account to the final destination node to house the results. Adjust the Schedule Trigger frequency based on how often you perform keyword research for your clients. Requirements SerpApi Account:** To programmatically access Google’s "People Also Ask" data. AI Credentials:** An API key for Gemini or OpenAI to generate the response text. Destination App:** A Notion workspace, Google Sheet, or Airtable base to store the output. Customising this workflow Competitor Tracking:** Modify the SerpApi parameters to include a specific domain to see what questions your competitors are currently ranking for. Schema Generation:** Add a second AI node to automatically wrap the answers in JSON-LD Question and AcceptedAnswer schema code for immediate dev-handoff. Multi-lingual AEO:** Add a translation node before the final output to localise answer snippets for international SEO campaigns. Slack Notification:** Add a Slack node at the end to notify the account manager to review the new optimisation opportunities. AEO Master Log:** Maintain a database to store all Q&A pairs for each client.

Converts the body of the email to markdown. Set email fields

This workflow automates the conversion of 2D images into high-quality 3D models (.glb format) by integrating *Google Sheets* with the *Hunyuan3D v2* model on *Fal.ai*. It handles the entire pipeline—from fetching image URLs to polling for completion and saving the final asset—eliminating manual modeling time for artists and developers. How it works This template operates on a schedule to process images in batches or individually: Data Retrieval: The workflow fetches new rows from a Google Sheet where the RESULT_GLB column is empty. AI Generation: It sends the IMAGE_URL to the Hunyuan3D v2 API on Fal.ai to initiate the 3D generation process. Status Polling: The workflow automatically enters a loop, checking the job status every 30 seconds until the model is marked "COMPLETED." Result Update: Once finished, it retrieves the download link for the .glb file and writes it back to the specific row in your Google Sheet. Use Cases Game Development:** Rapidly create prototype props and assets from concept art. E-commerce:** Convert product photos into 3D models for web viewers. AR/VR:** Generate background assets for immersive environments from simple 2D inputs. Setup steps Google Sheet: Create a new sheet with two header columns: IMAGE_URL and RESULT_GLB. Add the images you want to convert in the first column. Fal.ai Credentials: Sign up at Fal.ai and generate an API Key. In n8n, create a Header Auth credential with the name Authorization and value Key YOUR_API_KEY. Configure Nodes: Update the Get new image and Update Result nodes to select your specific Google Sheet. Ensure the HTTP Request nodes are using your Fal.ai Header Auth credential.

How it works This is an example of using sub-workflow nodes and a proof of concept showing that it’s possible to solve and explain recursive problems with n8n. Towers of Hanoi - Task Move a stack of n disks from rod A to rod C, using rod B as auxiliary. Only one disk can be moved at a time, and no disk may be placed on a smaller disk. Example n=4 | | | = | | === | | ===== | | ======= | | --------- --------- --------- A B C Algorithm procedure Hanoi(n, X, Y, Z): if n == 1: move disk from X to Z else: Hanoi(n-1, X, Z, Y) move disk from X to Z Hanoi(n-1, Y, X, Z) Notes This is a learning example. In a real scenario, you would probably use an iterative approach with only a single code node. When experimenting with recursion, make sure to define a termination condition first. Also, be aware of the "Restart workspace" link in the n8n Dashboard. Learn more about Recursion on Wikipedia&oldid=1301600240#Towers_of_Hanoi). Set up steps Optional: Set "numberOfDiscs" in node "Set number of discs" Execute workflow

No description available

WhatsApp Audio Transcriber Bot Overview Automatically transcribe WhatsApp audio messages to text using AI-powered speech recognition. This workflow receives audio messages via webhook, processes them through Groq's Whisper API, and replies with the transcribed text in the same conversation. Use Cases Accessibility**: Help users with hearing impairments access audio content Workplace Communication**: Quickly scan audio messages in professional settings Language Learning**: Get text versions of audio for better comprehension Meeting Notes**: Convert voice messages to searchable text format Multilingual Support**: Transcribe audio in Portuguese (configurable for other languages) How it Works Message Reception: Webhook receives WhatsApp messages in real-time Audio Detection: Filters only audio messages using Switch node Format Conversion: Converts base64 audio to MP3 file format AI Transcription: Processes audio through Groq API with Whisper Large V3 model Response Delivery: Sends transcribed text back to the original conversation Key Features ✅ Real-time Processing: Instant transcription of incoming audio messages ✅ High Accuracy: Uses Whisper Large V3 model for reliable transcription ✅ Auto-Reply: Automatically responds in the same WhatsApp conversation ✅ Message Quoting: References the original audio message in the reply ✅ Portuguese Optimized: Configured for Brazilian Portuguese transcription ✅ Self-Message Filtering: Ignores messages sent by the bot itself Prerequisites Required Services Evolution API**: WhatsApp integration service Groq API**: AI transcription service (Whisper model) n8n Instance**: Workflow automation platform API Keys & Configuration Groq API key (set as environment variable: GROQ_API_KEY) Evolution API instance properly configured Webhook URL configured in Evolution API Setup Instructions Import Workflow: Import the JSON workflow into your n8n instance Configure Environment: Set GROQ_API_KEY environment variable Setup Webhook: Configure Evolution API to send messages to the webhook endpoint Test Connection: Send a test audio message to verify the workflow Workflow Nodes Webhook**: Receives WhatsApp messages from Evolution API Edit Fields**: Extracts relevant data (number, name, message, audio) Switch**: Filters only audio messages (audioMessage type) Convert to File**: Transforms base64 audio to MP3 format HTTP Request**: Sends audio to Groq API for transcription Evolution API**: Sends transcribed text back to WhatsApp Configuration Options Groq API Settings Model**: whisper-large-v3 Language**: pt (Portuguese) Temperature**: 0 (maximum accuracy) Response Format**: json Customization Options Change language by modifying the language parameter Adjust temperature for different accuracy/creativity balance Modify response format for different output styles Response Format Mensagem transcrita automaticamente. [Transcribed text content] Technical Specifications Input**: Base64 encoded audio from WhatsApp Output**: Plain text transcription Processing Time**: Typically 2-5 seconds per audio message Supported Audio**: MP3 format (converted from WhatsApp audio) Language**: Portuguese (configurable) Troubleshooting No Response**: Check Groq API key and webhook configuration Poor Transcription**: Ensure audio quality and check language settings Error Messages**: Monitor n8n execution logs for detailed error information Version History v0.0.1**: Initial release with basic transcription functionality

This n8n workflow turns Telegram into a personal language tutor. Users can choose between different learning modes — vocabulary, grammar, quiz, or mixed lessons — simply by adding a hashtag to their message. The bot processes requests with AI/ML API and sends back structured, interactive lessons in Telegram. 🚀 Features 📩 Telegram-based input with hashtag commands 🧠 Adaptive AI responses (vocabulary, grammar, quiz) 🔤 Pronunciation support in Latin transcription 📒 Grammar explanations with examples ❓ Custom quizzes with auto-feedback 💬 Supportive learning experience with motivational messages ⏳ Typing indicator for smoother UX 🛠 Setup Guide 📲 Create Telegram Bot Go to @BotFather Use /newbot → choose a name and username Save the bot token 🔐 Set Up Credentials in n8n Telegram API: Use your bot token AI/ML API: Add your API key under AI/ML account ⚙️ Configure the Workflow Import the JSON into n8n Update credentials (Telegram + AI/ML API) Activate the workflow 💬 Start Learning In Telegram, send a message with one of the supported hashtags: #vocabulary — learn new words #grammar — study rules with examples #quiz — get exercises Or just send plain text for a free-form AI response 🔍 Node Overview Telegram Trigger** → Listens for incoming messages Show Typing Indicator** → Displays “typing…” while processing Switch Node** → Routes message by hashtag (#vocabulary, #grammar, #quiz) Prompt Builder Nodes** → Create JSON payload for AI/ML API AI/ML API Node** → Generates the structured lesson Telegram Send** → Sends the formatted response back to the user 💡 Example Flow User Input: #vocabulary кукуруза Bot Output: Кукуруза (Kukurúza) — Corn Pronunciation: koo-koo-ROO-zah Sentence: Я люблю есть кукурузу на гриле. I love eating grilled corn. 👉 Try to write your own sentence with “кукуруза”!

> ⚠️ Notice: > This workflow uses the CustomJS JSON to TOON node from CustomJS, which requires a self-hosted n8n instance and a CustomJS API key. This workflow demonstrates how to automatically categorize invoices from Airtable using OpenAI, while significantly reducing token usage by converting structured data into TOON (Token-Optimized Object Notation) before sending it to the LLM. Instead of passing large, nested JSON payloads (clients, invoices, invoice items) directly to OpenAI, the workflow converts the data into TOON, lets the AI focus purely on semantic classification, and then converts the result back into clean JSON that can safely be written back to Airtable. The result is a reliable, cost-efficient, and production-ready invoice categorization pipeline. Why This Workflow? Categorizing invoices based on free-text invoice items is a perfect AI use case — but sending raw JSON to LLMs is: expensive slow error-prone (broken JSON, hallucinated fields) This workflow applies the TOON pattern, reducing token usage by 40–60%, while ensuring the AI only returns exactly the data you expect. The LLM never sees raw JSON syntax — only a compact, schema-preserving representation. What This Workflow Does Fetches Invoices, Invoice Items, and Clients from Airtable Aggregates invoice items per invoice Converts the structured data into TOON Uses OpenAI to assign exactly one category per invoice Converts the AI result back into JSON Writes the category back to Airtable All without manual intervention. Key Features AI-based Invoice Categorization** Automatically classify invoices (e.g. web_development, system_administration) TOON Token Optimization** Reduce OpenAI token usage by 40–60% Schema Safety** AI never breaks your JSON structure Deterministic Output** One category per invoice, enforced by prompt design Airtable-Native** Reads from and writes directly back to Airtable How It Works Manual Trigger Run the workflow on demand Load Data from Airtable Fetch ready invoices Resolve related clients Fetch and aggregate invoice items Prepare Structured Context Group client, invoice, and invoice items into a single object JSON → TOON Convert structured data into token-efficient TOON AI Categorization OpenAI receives TOON data Assigns exactly one category per invoice Returns only TOON (no JSON, no prose) TOON → JSON Convert AI output back into valid JSON Persist Result Update the invoice category field in Airtable

Tagged with: , , , ,