Workflow Templates

This workflow will allow you to use OpenAI Assistant API together with a chatting platform. This version is configured to work with Hubspot, however, the Hubspot modules can be replaced by other platform and it will work similarly. Prerequisites: Create a Hubspot Chat (Live chat available on free plan) or Chatflow (paid hubspot only) and configure it to send all replies toward an n8n webhook (you need to create a custom app for that. I will create a separate article on how to do it, meanwhile, feel free to message me if you need support. Setup: Create a OpenAI Assistant, define its functionality and functions Update the Hubspot modules with the Hubspot API Key Update the OpenAI modules with OpenAI API Key Create an Airtable or any other database where you keep a reference between the thread id in Hubspot and Assistant API If you need help deploying this solution don't hesitate to email me or schedule a call here.

Automatically monitor LinkedIn posts from your community members and create AI-powered content digests for efficient social media curation. This template is perfect for community managers, content creators, and social media teams who need to track LinkedIn activity from their network without spending hours manually checking profiles. It fetches recent posts, extracts key information, and creates digestible summaries using AI. Good to know API costs apply** - LinkedIn API calls ($0.01-0.05 per profile check) and OpenAI processing ($0.001-0.01 per post) Rate limiting included** - Built-in random delays prevent API throttling issues Flexible scheduling** - Easy to switch from daily schedule to webhook triggers for real-time processing Requires API setup** - Need RapidAPI access for LinkedIn data and OpenAI for content processing How it works Daily profile scanning** - Automatically checks each LinkedIn profile in your Airtable for posts from yesterday Smart data extraction** - Pulls post content, engagement metrics, author information, and timestamps AI-powered summarization** - Creates 30-character previews of posts for quick content scanning Duplicate prevention** - Checks existing records to avoid storing the same post multiple times Structured storage** - Saves all processed data to Airtable with clean formatting and metadata Batch processing** - Handles multiple profiles efficiently with proper error handling and delays How to use Set up Airtable base** - Create tables for LinkedIn profiles and processed posts using the provided structure Configure API credentials** - Add your RapidAPI LinkedIn access and OpenAI API key to n8n credentials Import LinkedIn profiles** - Add community members' LinkedIn URLs and URNs to your profiles table Test the workflow** - Run manually with a few profiles to ensure everything works correctly Activate schedule** - Enable daily automation or switch to webhook triggers for real-time processing Requirements Airtable account** - For storing profile lists and managing processed posts with proper field structure RapidAPI Professional Network Data API** - Access to LinkedIn post data (requires subscription) OpenAI API account** - For intelligent content summarization and preview generation LinkedIn profile URNs** - Properly formatted LinkedIn profile identifiers for API calls Customising this workflow Change monitoring frequency** - Switch from daily to hourly checks or use webhook triggers for real-time updates Expand data extraction** - Add company information, hashtag analysis, or engagement trending Integrate notification systems** - Add Slack, email, or Discord alerts for high-engagement posts Connect content tools** - Link to Buffer, Hootsuite, or other social media management platforms for direct publishing Add filtering logic** - Set up conditions to only process posts with minimum engagement thresholds Scale with multiple communities** - Duplicate workflow for different LinkedIn communities or industry segments

Generate AI videos with Seedance & Blotato, upload to TikTok, YouTube & Instagram Who is this for? This template is ideal for creators, content marketers, social media managers, and AI enthusiasts who want to automate the production of short-form, visually captivating videos for platforms like TikTok, YouTube Shorts, and Instagram Reels — all without manual editing or publishing. What problem is this workflow solving? Creating engaging videos requires: Generating creative ideas Writing detailed scene prompts Producing realistic video clips and sound effects Editing and stitching the final video Publishing across multiple platforms This workflow automates the entire process, saving hours of manual work and ensuring consistent, AI-driven content output ready for social distribution. What this workflow does This end-to-end AI video automation workflow: Generates a creative idea using OpenAI and LangChain Creates detailed video prompts with Seedance AI Generates video clips via Wavespeed AI Generates sound effects with Fal AI Stitches the final video using Fal AI’s ffmpeg API Logs metadata and video links to Google Sheets Uploads the video to Blotato Auto-publishes to TikTok, YouTube, Instagram, and other platforms Setup Add your OpenAI API key in the LLM nodes Set up Seedance and Wavespeed AI credentials for video prompt and clip generation Add your Fal AI API key for sound and stitching steps Connect your Google Sheets account for tracking ideas and outputs Set your Blotato API key and fill in the platform account IDs in the Assign Social Media IDs node Adjust the Schedule Trigger to control when the automation runs How to customize this workflow to your needs Change the AI prompts** to target your niche (e.g., ASMR, product videos, humor) Add a Telegram or Slack step** for video preview before publishing Tweak scene structure** or video duration to match your style Disable platforms** you don’t want by turning off specific HTTP Request nodes Edit the sound generation prompts** for different moods or effects 📄 Documentation: Notion Guide Need help customizing? Contact me for consulting and support : Linkedin / Youtube

⚠️ Self-hosted only — This template uses a community node (Firecrawl) and cannot run on n8n Cloud. > 📸 Add a workflow screenshot at the top of this listing. Who it's for This workflow is for sales teams, lead generation agencies, and growth operators who maintain a company database in Airtable and need to automatically enrich records with phone numbers from multiple web sources. How it works A manual trigger fetches all Airtable records where the phone number field is empty. Records are processed one by one through a loop. Each company's website is scraped using Firecrawl, and the raw markdown is cleaned with a custom code node. A Claude Sonnet AI agent analyzes the cleaned content and attempts to extract a phone number using a structured output parser. If the AI finds no phone, the workflow falls back to an Apify LinkedIn scraper, first checking whether the company profile exists. If LinkedIn also yields nothing, a second fallback calls an Apify Google Maps scraper to search by company name and location. Whichever source returns a phone number, a code node normalizes the data and writes it back to Airtable. If no source finds a phone, the record is marked as unavailable. How to set up [ ] Connect your Airtable credentials and set the correct base and table in the Search Airtable Records, Update Airtable Phone Found, and Update Airtable Phone Not Found nodes [ ] Add your Firecrawl API key to the Scrape Company Website node [ ] Add your Anthropic API key to the Claude Sonnet Model sub-node inside AI Phone Finder Agent [ ] Add your Apify API token to both the Fetch LinkedIn Phone via Apify and Fetch Google Maps Phone via Apify nodes [ ] Review the Clean Scraped Markdown and Refine and Merge Phone Data code nodes to ensure field names match your Airtable schema Requirements Airtable account with a company database Firecrawl API key Anthropic API key (Claude Sonnet) Apify account with access to the LinkedIn Company Scraper and Google Maps Scraper actors How to customize Swap out either fallback source (LinkedIn or Google Maps) by bypassing the corresponding branch if you only have access to one Apify actor. Tune the AI system prompt inside the Claude Sonnet model to extract other contact details such as email addresses or physical addresses in addition to phone numbers. Replace the manual trigger with a scheduled trigger to run the enrichment automatically on a daily or weekly basis.

This n8n workflow automates the enrichment of a company list by discovering and extracting each company’s official LinkedIn URL using Bright Data’s search capabilities and Google Gemini AI for HTML parsing and result interpretation. Who is this template for? This workflow is ideal for sales, business development, and data research professionals who need to collect official LinkedIn company profiles for multiple organizations, starting from a list of company names in Google Sheets. It’s especially useful for teams who want to automate sourcing LinkedIn URLs, enrich their prospect database, or validate company data at scale. How it works Manual Trigger: The workflow is started manually (useful for controlled batch runs and testing). Read Company Names: Company names are loaded from a specified Google Sheets table. Loop Over Each Company: Each company is processed one-by-one: A custom Google Search URL is generated for each name. A Bright Data Web Unlocker request is sent to fetch Google search results for “site:linkedin.com [company name]”. Parse LinkedIn Profile URL Using AI: Google Gemini (or your specified LLM) analyzes the fetched search page and extracts the most likely official LinkedIn company profile. Result Handling: If a profile is found, it’s stored in the results. If not, an empty result is created, but you can add custom logic (notifications, retries, etc.). Batch Data Enrichment: All found company URLs are bundled into a single request for further enrichment from a Bright Data dataset. Export: The workflow appends the final, structured data for each company to another sheet in your Google Sheets file. Setup instructions 1. Replace API Keys: Insert your Bright Data API key in these nodes: Bright Data Web Request - Google Search for Company LinkedIn URL HTTP Request - Post API call to Bright Data Snapshot Progress HTTP Request - Getting data from Bright Data 2. Connect Google Sheets: Set up your Google Sheets credentials and specify the sheet for reading input and writing output. 3. Customize Output Structure: Adjust the Python code node (see sticky note in the template) if you want to include additional or fewer fields in your output. 4. Adjust for Scale or Error Handling: You can modify the logic for “not found” results (e.g., to notify a Slack channel or retry failed companies). 5. Run the Workflow: Start manually, monitor the run, and check your Google Sheet for results. Customization guidance Change Input/Output Sheets: Update the sheet names or columns if your source/target spreadsheet has a different structure. Use Another AI Model: Replace the Google Gemini node with another LLM node if preferred. Integrate Alerts: Add Slack or email nodes to notify your team when a LinkedIn profile is not found or when the process is complete.

This template demonstrates how to provide an interactive chatbot for your work history based off your CV. Unanswered questions and follow-up email contacts are sent to you via Telegram. Use case: link on your profile to not only show off you AI workflow skills but also to provide an interactive chatbot about your work history for prospective employers. Good to Know It will require access to an OpenAI API Key (free for low usage) and setting up a bot in Telegram (free). How it Works The n8n inbuilt chat node will be hosted on n8n services to provide the chat interface. You will upload your CV either exported from LinkedIn or exported yourself to Microsoft OneDrive along with a simple text file explaining some other information about you. On each chat interaction the PDF and text file are used as tools to get context information for the chatbot to respond. If a question cannot be answered reliably, a subworkflow will be called to capture that question and send it to you as a telegram message. If the person chatting supplies their email address, this will be sent to you via a Telegram message along with other information the user provides. How to use After importing the template, create the subworkflows so that they can be used a Tools by the AI Node. Don't forget to add the Execute sub-workflow trigger. Setup credentials for Open AI, OneDrive and telegram. Upload your CV & text file summary to OneDrive and replace the document IDs in the get_documents sub-workflow. Activate the workflow so that publicly available chat will get generated on n8n.

Who’s it for This template is perfect for OMI pendant users or anyone with AI-generated memory transcripts who want to: Automatically create daily journals in Markdown Extract actionable tasks from conversations Store memories in Google Drive Sync action items to Google Tasks Great for creators, ADHD professionals, techies, or productivity hackers who want to build a second brain workflow with no manual data entry. What it does / How it works This workflow: Accepts POST data from the OMI AI pendant (via webhook) Extracts structured summaries, tasks, events, and raw transcript data Converts the transcript into Markdown using metadata like emoji, category, and overview Uses Google Gemini or an AI Agent to generate a high-quality journal entry Splits out action items and creates tasks in Google Tasks Uploads both the transcription and the final journal file into separate Google Drive folders for archival Deletes processed files if needed (cleanup path is included) How to set up Connect your OMI device to send daily summaries to the webhook endpoint Authenticate your Google Drive and Google Tasks accounts Replace any hardcoded values (like folder IDs or task list IDs) with your own Review the system prompt in the AI Agent node if you'd like to personalize your journal style ## Requirements OMI pendant or device that generates .md summaries via API or webhook Google Drive & Google Tasks credentials set up in n8n Optional: Google Gemini or OpenAI for natural language journal generation ## How to customize Change the output folder IDs for GDrive in the Upload Transcription and Upload Journal nodes. One folder is for long term storage and the other is short term, the contents of which are deleted every night to generate the journal entries. Ensure your workflow timezone is set correctly in the settings. Replace Google Tasks with another todo app (e.g. Notion, Todoist) using HTTP or native nodes Customize the AI prompt in the AI Agent or Gemini Chat node to reflect your tone (e.g., poetic, minimalist, spiritual) Modify the Markdown format in the Build Markdown Transcription node for your preferred structure

Empower Your AI Chatbot with Long-Term Memory and Dynamic Tool Routing This n8n workflow equips your AI agent with long-term memory and a dynamic tools router, enabling it to provide intelligent, context-aware responses while managing tasks across multiple tools. By combining persistent memory and modular task routing, this workflow makes your AI smarter, more efficient, and highly adaptable. 👥 Who Is This For? AI Developers & Automation Enthusiasts: Integrate advanced AI features like long-term memory and task routing without coding expertise. Businesses & Teams: Automate tasks while maintaining personalized, context-aware interactions. Customer Support Teams: Improve user experience with chatbots that remember past interactions. Marketers & Content Creators: Streamline communication across platforms like Gmail and Telegram. AI Researchers: Experiment with persistent memory and multi-tool integration. 🚀 What Problem Does This Solve? This workflow simplifies the creation of intelligent AI systems that retain memory, manage tasks dynamically, and automate notifications across tools like Gmail and Telegram—saving time and improving efficiency. 🛠️ What This Workflow Does Save & Retrieve Memories**: Uses Google Docs for long-term storage to recall past interactions or user preferences. Dynamic Task Routing**: Routes tasks to the right tools (e.g., saving/retrieving memories or sending notifications). AI-Powered Context Understanding**: Combines OpenAI GPT-based short-term memory with long-term memory for smarter responses. Multi-Channel Notifications**: Sends updates via Gmail or Telegram. 🔧 Setup API Credentials: Connect to OpenAI (AI processing), Google Docs (memory storage), Gmail/Telegram (notifications). Customize Parameters: Adjust the AI agent's system message for your use case. Define task-routing rules in the tools router node. Test & Deploy: Verify memory saving/retrieval, task routing, and notification delivery. 💡 How to Customize Modify the system message in the OpenAI node to tailor your agent’s behavior. Add or adjust routing rules for additional tools. Update notification settings to match your communication preferences.

This n8n workflow is a Telegram bot that allows users to either: Generate AI images using Pollinations API, or Generate blog articles using Gemini AI Users simply type image your prompt or blog your title, and the bot responds with either an AI-generated image or article. Who's it for This template is ideal for: Content creators and marketers who want to generate visual and written content quickly Telegram bot developers looking for real-world AI integration Educators or students automating content workflows Anyone managing content pipelines using Google Sheets What it does / How it works Telegram Interaction Trigger Telegram Message: Listens for new messages or button clicks via Telegram Classify Telegram Input: JavaScript logic to classify input as /start, /help, normal text, or callback Switch Input Type: Directs the flow based on the classification Menu & Help Send Main Menu to User: Shows "Generate Image", "Blog Article", "Help" options Switch Callback Selection: Routes based on button pressed (image, blog, or help) Send Help Instructions: Sends markdown instructions on how to use the bot Input Validation Validate Command Format: Ensures input starts with image or blog Notify Invalid Input Format: If validation fails, informs user of correct format Image Generator Prompt User for Image Description → When user clicks Generate Image Detect Text-Based Input Type → Detects if text is image or blog Switch Text Command Type → Directs whether to generate image or article Show Typing for Image Generation → Sends "uploading photo..." typing status Build Image Generation URL → Constructs Pollinations API image URL from prompt Download AI Image → Makes HTTP request to get the image Send Image Result to Telegram → Sends image to user via Telegram Log Image Prompt to Google Sheets → Logs prompt, image URL, date, and user ID Upload Image to Google Drive → Saves image to Google Drive folder Blog Article Generator Prompt User for Blog Title → When user clicks Blog Article Store Blog Prompt → Saves prompt for later use Log Blog Prompt to Google Sheets → Writes title + user ID to Google Sheets Send Article Style Options → Offers: Formal, Casual, or News style Store Selected Article Style → Updates row with chosen style in Google Sheets Fetch Last User Prompt → Finds the latest prompt submitted by this user Extract Last Blog Prompt → Extracts row for use in AI request Gemini Chat Wrapper → Handles input into LangChain node for AI processing Generate Article with Gemini → Calls Gemini to create 3-paragraph blog post Parse Gemini Response → Parses JSON string to extract title and content Send Article to Telegram → Sends blog article result back to user Log Final Article to Google Sheets → Updates row with final content and timestamp Requirements Telegram bot (via @BotFather) Pollinations API (free and public endpoint) Google Sheets & Drive (OAuth credential setup in n8n) Google Gemini / PaLM API key via LangChain Self-hosted or cloud n8n setup Setup Instructions Clone the workflow and import it into your n8n instance Set credentials: Telegram API Google Sheets OAuth Google Drive OAuth Gemini (via LangChain) Replace: Sheet ID with your own Google Sheet Folder ID on Google Drive chat_id placeholders if needed (use expressions instead) Deploy and send /start in your Telegram bot 🔧 Customization Tips Edit the Gemini prompt to adjust article length or tone Add extra style buttons like "SEO", "Story", "Academic" Add image post-processing (e.g. compression, renaming) Add error catching logic (e.g. if Pollinations image fails) Store images with filenames based on timestamp/user Security Considerations Use n8n credentials for all tokens (Telegram, Gemini, Sheets, Drive) Never hardcode your token inside HTTP nodes Do not expose real Google Sheet or Drive links in shared version Use Set node to collect all editable variables (like folder ID, sheet name)

Multi-Agent Book Creation Workflow with AI Tool Node and GPT-4, DALL-E Who’s it for This workflow is designed for: Content creators** who want to generate books or structured documents automatically. Educators and trainers** who need quick course materials, eBooks, or study guides. Automation enthusiasts* exploring *multi-agent systems* using the newly released *AI Tool Node** in n8n. Developers* looking for a reference template to understand *orchestration of multiple AI agents** with structured output. How it works / What it does This template demonstrates a multi-agent orchestration system powered by AI Tool Nodes: Trigger: Workflow starts when a chat message is received. Book Brief Agent: Generates the initial book concept (title, subtitle, and outline). Book Writer Agent: Expands the outline into full content by collaborating with two sub-agents: Designer Agent → Provides layout/design suggestions. Content Writer Agent → Drafts and refines chapters. Generate Cover Image: AI generates a custom book cover image. Upload to AWS S3: Stores the cover image securely. Configure Metadata: Adds metadata for title, author, and description. Build Book HTML: Converts markdown-based content into HTML format. Upload to Google Drive: Saves the HTML content for processing. Convert to PDF: Transforms the book into a professional PDF. Archive to Google Drive: Final version is archived for safe storage. This workflow showcases multi-agent coordination, structured parsing, and seamless integration with cloud storage services. How to set up Import the workflow into n8n. Configure the following connections: OpenAI (for Book Brief, Book Writer, Designer, and Content Writer Agents). AWS S3 (for image storage). Google Drive (for document storage & archiving). Add your API keys and credentials in n8n credentials manager. Test the workflow by sending a sample chat message (e.g., “Write a book about AI in education”). Verify outputs in Google Drive (HTML + PDF) and AWS S3 (cover image). Requirements n8n* (latest version with *AI Tool Node** support). OpenAI API key** (to power multi-agent models). AWS account** (with S3 bucket for storing images). Google Drive integration** (for document storage and archiving). Basic familiarity with workflow setup in n8n. How to customize the workflow Switch Models**: Replace gpt-4.1-mini with other models (faster, cheaper, or more powerful). Add More Agents: Introduce agents for **editing, fact-checking, or translation. Change Output Format: Export to **EPUB, DOCX, or Markdown instead of PDF. Branding Options: Modify the **cover generation prompt to include company logos or specific style. Extend Storage: Add **Dropbox, OneDrive, or Notion integration for additional archiving. Trigger Alternatives: Replace chat trigger with **form submission, webhook, or schedule-based runs. ✅ This workflow acts as a free, plug-and-play template to showcase how multi-agents + AI Tool Node can work together to automate complex content creation pipelines.

What this template does The Ultimate Scraper for n8n uses Selenium and AI to retrieve any information displayed on a webpage. You can also use session cookies to log in to the targeted webpage for more advanced scraping needs. ⚠️ Important: This project requires specific setup instructions. Please follow the guidelines provided in the GitHub repository: n8n Ultimate Scraper Setup : https://github.com/Touxan/n8n-ultimate-scraper/tree/main. The workflow version on n8n and the GitHub project may differ; however, the most up-to-date version will always be the one available on the GitHub repository : https://github.com/Touxan/n8n-ultimate-scraper/tree/main. How to use Deploy the project with all the requirements and request your webhook. Example of request: curl -X POST http://localhost:5678/webhook-test/yourwebhookid \ -H "Content-Type: application/json" \ -d '{ "subject": "Hugging Face", "Url": "github.com", "Target data": [ { "DataName": "Followers", "description": "The number of followers of the GitHub page" }, { "DataName": "Total Stars", "description": "The total numbers of stars on the different repos" } ], "cookie": [] }' Or to just scrap a url : curl -X POST http://localhost:5678/webhook-test/67d77918-2d5b-48c1-ae73-2004b32125f0 \ -H "Content-Type: application/json" \ -d '{ "Target Url": "https://github.com", "Target data": [ { "DataName": "Followers", "description": "The number of followers of the GitHub page" }, { "DataName": "Total Stars", "description": "The total numbers of stars on the different repo" } ], "cookies": [] }' `

Instructions This automation enables you to just upload any Image (via Form) of a Logo Sheet, containing multiple Images of Product Logos (most likely) which brings them in some context to one another. After submitting an AI-Agent eats that Logo Sheet, turning it into an List of "Productname" and "Attributes", also checks if Tools are kind of similar to another, given the Context of the Image. We utilize AI Vision capabilities for that. NOTE: It might not be able to extract all informations. For a "upload and forget it" Workflow it works for me. You can even run it multiple times, to be sure. But if you need to make sure it extracts everything you might need to think about an Multi-Agent Setup with Validation-Agent Steps. Once the Agent finishes the extraction, it will traditionally and deterministicly add those Attributes to Airtable (Creates those, if not already existing.) and also Upserts the Tool Informations. It uses MD5 Hashes for turning Product Names into.. something fancy really, you could also use it without that, but I wanted to have something that looks atleast like an ID. Setup Set Up the Airtable like shown below. Update and set Credentials for all Airtable Nodes. Check or Adjust the Prompt of the Agent matching your use-case. Activate the Workflow. Open the Form (default: https://your-n8n.io/form/logo-sheet-feeder) Enjoy growing your Airtable. Enjoy the workflow! ❤️ let the work flow — Workflow Automation & Development