Agents & Automations

Search music and play to Spotify from Telegram. This workflow is a simple demonstration on accessing a message model from Telegram, making searching for songs an easy task even if you can't remember the artist or song name. An OpenAI message model tries to figure out the song and sends it to an active Spotify device. Use case Imagine an office where you play music in the background and the employees can control the music without having to log in to the playing account. How it works 1. You describe the song in Telegram. 2. The Telegram bot sends the text to n8n. 3. An OpenAI message model tries to find the song. 4. Spotify gets the search query string. 5. The first match is then added to the queue. - If there is no match, a message is sent to Telegram and the process ends. 6. We change to the next track in the list. 7. We make sure the song starts playing by trying to resume. 8. We fetch the currently playing track. 9. We return "now playing" information to Telegram: Song Name - Artist Name - Album Name. Error handling Every Spotify step has its own error handler under settings where we output the error. The message parser receives the error and sends it to Telegram. Requirements - Active workflow - OpenAI API key - Telegram bot - Spotify account and Oauth2 API - Spotify active on a device The Telegram trigger is activated only if this workflow is active. You can, however, TEST the workflow in the editor by clicking "Test step" and then it waits for the Telegram event. When the event is received, just step through all steps or click "Test step" on the "Fetch Now Playing" node. You must have a Spotify device active when trying to communicate with a device. Open Spotify and play something — not it is active.

Video Guide I prepared a comprehensive guide detailing how to automate the parsing of invoices using n8n and LlamaParse, seamlessly capturing and storing vital billing information. Who is this for? This workflow is ideal for finance teams, accountants, and business operations managers who need to streamline invoice processing. It is particularly helpful for organizations seeking to reduce manual entry errors and improve efficiency in managing billing information. What problem does this workflow solve? Manually processing invoices can be time-consuming and error-prone. This automation eliminates the need for manual data entry by capturing invoice details directly from uploaded documents and storing structured data efficiently. This enhances productivity and accuracy across financial operations. What this workflow does The workflow leverages n8n and LlamaParse to automatically detect new invoices in a designated Google Drive folder, parse essential billing details, and store the extracted data in a structured format. The key functionalities include: - Real-time detection of new invoices via Google Drive triggers. - Automated HTTP requests to initiate parsing through Llama Cloud. - Structured storage of invoice details and line items in a database for future reference. - Google Drive Integration: Monitors a specific folder in Google Drive for new invoice uploads. - Parsing with LlamaParse: Automatically sends invoices for parsing and processes results through webhooks. - Data Storage in Airtable: Creates records for invoices and their associated line items, allowing for detailed tracking. SetupN8N Workflow 1. Google Drive Trigger: Set up a trigger to detect new files in a specified folder dedicated to invoices. 2. File Upload to LlamaParse: Create an HTTP request that sends the invoice file to LlamaParse for parsing, including relevant header settings and webhook URL. 3. Webhook Processing: Establish a webhook node to handle parsed results from LlamaParse, extracting needed invoice details effectively. 4. Invoice Record Creation: Create initial records for invoices in your database using the parsed details received from the webhook. 5. Line Item Processing: Transform string data into structured line item arrays and create individual records for each item linked to the main invoice.

This n8n workflow demonstrates how to create a really simple yet effective customer support channel and pipeline by combining Slack, Linear, and AI tools. Built on n8n's ability to integrate anything, this workflow is intended for small support teams who want to maximize the re-use of the tools they already have with an interface that doesn't require any onboarding. How it works The workflow is connected to a Slack channel set up with the customer to capture support issues. Only messages which are tagged with a "✅" reaction are captured by the workflow. Messages are tagged by the support team in the channel. Each captured support issue is sent to the AI model to classify, prioritize, and rewrite into a support ticket. The generated support ticket is uploaded to Linear for the support team to investigate and track. The support team is able to report back to the user via the channel when the issue is fixed. Requirements - Slack channel to be monitored - Linear account and project Customizing this workflow Don't have Linear? This workflow can work just as well with traditional ticketing systems like JIRA.

This n8n demonstrates how to build a simple YouTube MCP server to look up videos on YouTube and download their transcripts for research purposes. YouTube videos are a great source of new and updated information on a variety of cutting-edge developments, but they're not always simple to understand, and lengthy videos may take too much time. Using this MCP server, you can extract and feed their transcripts for your AI agents, which then allows it to break down the content into manageable learnings and insights. How it works A MCP server trigger is used and connected to three custom workflow tools: YouTube Search, YouTube Transcripts, and Usage Reports. Both YouTube tools use an external scraping service called APIFY.com. This is my preference as it's a much simpler interface and there are no rate limits. The YouTube Search fetches 10 results based on the user's query. The YouTube Transcripts downloads the subtitles from one or more given URLs. The usage reports pull in your monthly APIFY.com spending and limits as a way to check your account. How to use This Apify YouTube MCP server allows any compatible MCP client to research YouTube videos for any desired topic. An Apify account is required, however, to connect and use the service. Connect your MCP client by following the n8n guidelines. Alternatively, connect any n8n AI agent with the MCP client tool. Try the following queries in your MCP client: - "What is MCP?" - "How can I use MCP in n8n?" - "How can I use Apify's official MCP server?" Requirements - APIFY.com for YouTube Scraping. This is a paid service, but there is a $5 free tier which is ample for this template. - MCP Client or Agent for usage such as Claude Desktop. Customizing this workflow Add as many APIFY.com actors as required for your use case or users. Consider using Apify's official MCP server for 4000+ available tools. Remember to set the MCP server to require credentials before going to production and sharing this MCP server with others!

This n8n workflow template lets teams easily generate a custom AI chat assistant based on the schema of any Notion database. Simply provide the Notion database URL, and the workflow downloads the schema and creates a tailored AI assistant designed to interact with that specific database structure. Key FeaturesInstant Assistant Generation: Enter a Notion database URL, and the workflow produces an AI assistant configured to the database schema. Advanced Querying: The assistant performs flexible queries, filtering records by multiple fields (e.g., tags, names). It can also search inside Notion pages to pull relevant content from specific blocks. Schema Awareness: Understands and interacts with various Notion column types like text, dates, and tags for accurate responses. Reference Links: Each query returns direct links to the exact Notion pages that inform the assistant’s response, promoting transparency and easy access. Self-Validation: The workflow has logic to check the generated assistant, and if any errors are detected, it reruns the agent to fix them. Ideal for - Product Managers: Easily access and query product data across Notion databases. - Support Teams: Quickly search through knowledge bases for precise information to enhance support accuracy. - Operations Teams: Streamline access to HR, finance, or logistics data for fast, efficient retrieval. - Data Teams: Automate large dataset queries across multiple properties and records. How It Works This AI assistant leverages two HTTP request tools—one for querying the Notion database and another for retrieving data within individual pages. It’s powered by the Anthropic LLM (or can be swapped for GPT-4) and always provides reference links for added transparency.

When a new image is added to a monitored folder in Google Drive, Picsart automatically removes the background, then seamlessly uploads the enhanced image back to Google Drive. For any inquiries about the template, please reach out to the template creator: [email protected].

Easily transform your engaging TikTok videos into professional LinkedIn draft posts with ChatGPT, where you can review the content before posting. Simply email the video link and receive the content in your inbox. Significantly reduce the time from inspiring video you find to LinkedIn post content to 1 minute.

WebSecScan: AI-Powered Website Security Auditor This n8n workflow provides comprehensive website security analysis by leveraging OpenAI's models to detect vulnerabilities, configuration issues, and security misconfigurations. The workflow generates a professional HTML security report delivered directly via Gmail. Key Features - Dual-Layer Security Analysis: Performs parallel security audits using specialized OpenAI agents: - Header Configuration Audit: Analyzes HTTP headers, CORS policies, CSP implementation, and cookie security. - Vulnerability Assessment: Identifies XSS vectors, information disclosure, and client-side weaknesses. - Detailed Security Grading: Automatically calculates a security grade (A+ to F) based on findings severity and quantity. - Professional Report Generation: Creates a comprehensive HTML report with: - Security grade visualization - Color-coded vulnerability categories - Detailed recommendations with example configuration fixes - Header presence/absence indicators - Implementation guidance for remediation - Non-Invasive Testing: Performs analysis without active scanning or exploitation attempts. Technical Implementation - Multi-Agent Architecture: Utilizes two specialized OpenAI agents with custom prompts tailored for security analysis. - Advanced Header Analysis: Detects presence and proper implementation of critical security headers: - Content-Security-Policy - Strict-Transport-Security - X-Content-Type-Options - X-Frame-Options - Referrer-Policy - Permissions-Policy - Intelligent Issue Detection: Uses JavaScript processing to analyze OpenAI outputs and count critical/warning issues. - Responsive HTML Report: Dynamically generates a mobile-friendly report with detailed findings and recommendations. Setup Requirements 1. OpenAI API Configuration - Create an OpenAI API key at platform.openai.com. - In n8n, go to Settings → Credentials → New → OpenAI API. - Enter your API key and save. 2. Gmail Integration - Navigate to Settings → Credentials → New → Gmail OAuth2 API. - Complete the OAuth authentication flow. - Configure recipient email in the "Send Security Report" node. 3. Workflow Customization (Optional) - Modify the form title/description in the Landing Page node. - Upgrade from gpt-4o-mini to gpt-4o for more comprehensive analysis. - Add additional recipients to the email report. Usage Instructions - Activate the workflow and access the form via the generated URL. - Enter any website URL to analyze (including the http:// or https:// prefix). - Receive a detailed security report via email within minutes. - Share findings with your development team to implement fixes. This workflow represents a non-invasive security assessment tool. For production environments, complement with professional penetration testing services.

This n8n template leverages n8n's multi-form feature to build a 2-part job application submission journey which aims to eliminate the need for applicants to re-enter data found on their CVs/Resumes. How it works The application submission process starts with an n8n form trigger to accept CV files in the form of PDFs. The PDF is validated using the text classifier node to determine if it is a valid CV; else, the applicant is asked to reupload. A basic LLM node is used to extract relevant information from the CV as data capture. A copy of the original job post is included to ensure relevancy. Applicant's data is then sent to an ATS for processing. For our demo, we used Airtable because we could attach PDFs to rows. Finally, a second form trigger is used for the actual application form. However, it is prefilled to save the applicant's time and allow them to amend any of the generated application fields. How to use Ensure to change the redirect URL in the form ending node to use the host domain of your n8n instance. Requirements - OpenAI for LLM - Airtable to capture applicant data Customizing the workflow The application form is pretty basic for this demonstration but could be extended to ask more in-depth questions. If it fits the job, why not ask applicants to upload portfolio works and have AI describe/caption them?

Who is this for? This workflow is designed for: - Content creators, artists, or hobbyists looking to experiment with AI-generated art. - Small business owners or marketers using LEGO-style designs for branding or promotions. - Developers or AI enthusiasts wanting to automate image transformations through messaging platforms like LINE. What problem is this workflow solving? - Simplifies the process of creating custom AI-generated LEGO-style images. - Automates the manual effort of transforming user-uploaded images into AI-generated artwork. - Bridges the gap between messaging platforms (LINE) and advanced AI tools (DALL·E). - Provides a seamless system for users to upload an image and receive an AI-transformed output without technical expertise. What this workflow doesImage Upload via LINE: Users send an image to the LINE chatbot. AI-Powered Prompt Creation: GPT generates a prompt to describe the uploaded image for LEGO-style conversion. AI Image Generation: DALL·E 3 processes the prompt and creates a LEGO-style isometric image. Image Delivery: The generated image is returned to the user in LINE. SetupPrerequisites - LINE Developer Account with API credentials. - Access to OpenAI API with DALL·E and GPT-4 capabilities. - A configured n8n instance to run this workflow. StepsEnvironment Setup: Add your LINE API Token and OpenAI credentials as environment variables (LINE_API_TOKEN, OPENAI_API_KEY) in n8n. Configure LINE Webhook: Point the LINE webhook to your n8n instance. Connect OpenAI: Set up OpenAI API credentials in the workflow nodes for GPT-4 and DALL·E. Test Workflow: Upload a sample image in LINE and ensure it returns the LEGO-style AI image. How to customize this workflow to your needsLocalization: Modify response messages in LINE to fit your audience's language and tone. Integration: Add nodes to send notifications through other platforms like Slack or email. Image Style: Replace the LEGO-style image prompt with other artistic styles or themes. Advanced Use CasesArt Contests: Users upload images and receive AI-enhanced outputs for community voting or branding. Marketing Campaigns: Quickly generate creative visual content for ads and promotions using customer-submitted photos. Education: Use the workflow to teach students about AI-generated art and automation through a hands-on approach. Tips for OptimizationError Handling: Add fallback nodes to handle invalid images or API errors gracefully. Logging: Implement a logging mechanism to track requests and outputs for debugging and analytics. Scalability: Use queue-based systems or cloud scaling to handle large volumes of image requests. Enhancements - Add sticky notes in n8n to provide inline instructions for configuring each node. - Create a tutorial video or documentation for first-time users to set up and customize the workflow. - Include advanced filters to allow users to select from multiple styles beyond LEGO (e.g., pixel art, watercolor). This workflow enables seamless interaction between messaging platforms and advanced AI capabilities, making it highly versatile for various creative and business applications.

Imagine having an AI chatbot on Slack that seamlessly integrates with your company’s workflow, automating repetitive requests. No more digging through emails or documents to find answers about IT requests, company policies, or vacation days—just ask the bot, and it will instantly provide the right information. With its 24/7 availability, the chatbot ensures that team members get immediate support without waiting for a colleague to be online, making assistance faster and more efficient. Moreover, this AI-powered bot serves as a central hub for internal communication, allowing everyone to quickly access procedures, documents, and company knowledge without searching manually. A simple Slack message is all it takes to get the information you need, enhancing productivity and collaboration across teams. ### How It Works Slack Trigger: The workflow starts when a user mentions the AI bot in a Slack channel. The trigger captures the message and forwards it to the AI Agent. AI Agent Processing: The AI Agent, powered by Anthropic's Claude 3.7 Sonnet model, processes the query. It uses Retrieval-Augmented Generation (RAG) to fetch relevant information from the company’s internal knowledge base stored in Qdrant (a vector database). A Simple Memory buffer retains recent conversation context (last 10 messages) for continuity. Knowledge Retrieval: The RAG tool searches Qdrant’s vector store using OpenAI embeddings to find the most relevant document chunks (top 10 matches). Response Generation: The AI synthesizes the retrieved data into a concise, structured response (1-2 sentences for the answer, 2-3 supporting details, and a source citation). The response is formatted in Slack-friendly markdown (bullet points, blockquotes) and sent back to the user. ### Set Up Steps Prepare Qdrant Vector Database: - Create a Qdrant collection via HTTP request (Create collection node). - Optionally, refresh/clear the collection (Refresh collection node) before adding new documents. Load Company Documents: - Fetch files from a Google Drive folder (Get folder → Download Files). - Process documents: Split text into chunks (Token Splitter) and generate embeddings (Embeddings OpenAI2). - Store embeddings in Qdrant (Qdrant Vector Store1). Configure Slack Bot: - Create a Slack bot via Slack API with required permissions. - Add the bot to the desired Slack channel and note the channelId for the workflow. Deploy AI Components: - Connect the AI Agent to Anthropic’s model, RAG tool, and memory buffer. - Ensure OpenAI embeddings are configured for both RAG and document processing. Test & Activate: - Use the manual trigger (When clicking ‘Test workflow’) to validate document ingestion. - Activate the workflow to enable real-time Slack interactions. Need help customizing? Contact me for consulting and support or add me on LinkedIn.

About The Airtop Automation Are you tired of being shocked by unexpectedly high energy bills? With this automation using Airtop and n8n, you can take control of your daily energy costs and ensure you’re always informed. How to monitor your daily energy consumption With this automation, we’ll walk you through setting up an automation that retrieves your PG&E (Pacific Gas and Electric) energy usage data, calculates costs, and emails you the details—all without manual effort. What You’ll Need To get started, make sure you have the following: - A free Airtop API Key - PG&E Account Credentials - with minor adaptations, this will also work with other providers - An Email Address - To receive the energy cost updates Estimated setup time: 5 minutes Understanding the Process This automation works by: - Logging into your PG&E account using your credentials - Navigating to your energy usage data - Extracting relevant details about energy consumption and costs - Emailing the daily summary directly to your inbox The automation is straightforward and ensures you have real-time insights into your energy usage, empowering you to adjust your habits and save money. Setting Up Your Automation We’ve created a step-by-step guide to help you set up this workflow. Here’s how: 1. Insert Your Credentials: - In the tools section, add your PG&E login details as variables - In Airtop, add your Airtop API Key - Configure your email address to receive the updates 2. Run the Automation: - Start the scenario, and watch as the automation retrieves your energy data and sends you a detailed email summary. Customization Options While the default setup works seamlessly, you can tweak it to suit your needs: - Data Storage: Store energy usage data in a database for long-term tracking and analysis - Visualization: Plot graphs of your energy usage trends over time for better insights - Notifications: Change the automation to only send alerts on high usage instead of a daily email Real-World Applications This automation isn’t just about monitoring energy usage and taking control. Here are some practical applications: - Daily Energy Management: Receive updates every morning and adjust your energy consumption based on costs - Smart Home Integration: Use the data to automate appliances during off-peak hours - Budgeting: Track energy expenses over weeks or months to plan your budget more effectively Happy automating!

This template is based on the following framework. Thank you for the groundwork, Matheus. How it works: Store your snippets of text in a Notion table. Each snippet should have an image associated with it (copy + pasted into the text). Connect to your table via a Notion "integration," from which N8N can then query your pre-meditated posts. The text is fed through an OpenAI assistant to boost engagement via formatting. The re-formatted text, along with the image pulled from the Notion snippet, are combined into a post for your LinkedIn. The row in the original Notion table from the first step containing this post is set to a status of "Done." Set up steps: You will need to create a Notion "integration," which will yield a "secret key" that you enter into your N8N as a "Credential." You will need to create a LinkedIn "app" in order to post on your behalf. When creating your LinkedIn "app," you will be required to link this "app" to a company page on LinkedIn. If you are doing this for yourself, search for the "Default Company Page (for API testing)," and select this page as it is provided by LinkedIn for individuals. You can find your LinkedIn apps here, and if you get stuck, further instructions on setting up this workflow (including this LinkedIn OAuth piece) can be found in this YouTube Video Aide to these instructions. Lastly, you will need to create an OpenAI API key, found on your OpenAI Playground Dashboard. Once you created an API key, make sure you have an assistant created from the "Assistants" tab on the OpenAI dashboard. This assistant and its instructions will be needed for carrying out the re-formatting of your post.

Automate your email management with this workflow, designed for freelancers and business professionals who receive high volumes of emails. By leveraging AI-powered categorisation and dynamic email processing, this template helps you organise your inbox and streamline communication for better efficiency and productivity. Check out the YouTube video for step-by-step setup instructions! How it works:Fetch & Filter Emails: The workflow retrieves emails from your Microsoft Outlook account, filtering out flagged emails and those already categorised. Content Preparation: Each email is cleaned up and converted to a structured format using Markdown, making it easier for AI processing. AI Categorization: The content is analysed using an AI model, which categorises the emails into predefined categories (e.g., Action, Junk, Business, SaaS) based on the context and content. Email Categorization & Folder Management: The categorised emails are updated in Microsoft Outlook and moved to respective folders such as "Junk Email" or "Receipts" based on the AI's classification. Conditional Processing & Final Checks: Additional checks and conditions ensure that only unread emails are processed, and errors are gracefully managed to maintain workflow stability. Set up steps: 1. Connect Microsoft Outlook: Link your Microsoft Outlook account using the built-in credentials node to enable email fetching, updating, and folder management. 2. Configure AI Model (Ollama API): Set up the AI model by connecting to the Ollama API and choosing your desired language model for categorisation. 3. Modify Email Categories (Optional): Customize the categories and subcategories within the workflow to suit your unique email management needs. 4. Set Up Error Handling: Review the error handling node settings to ensure smooth workflow execution. This template offers a robust solution for managing and organising your inbox, helping you save time and keep your focus on important emails.

Every time a new image containing text is uploaded to your Google Drive folder, Make will detect the text with Google Cloud Vision (OCR), translate it into the language you want using Google Translate, and save the translated text as a new file to a Google Drive folder of your choice.

Quickly organize your inbox with AI! This simple workflow automatically classifies incoming emails into different categories — like High Priority, Work Related, or Promotions — and applies Gmail labels accordingly. Setup takes less than 2 minutes, and it runs 24/7, helping you stay focused on what matters most without manual sorting. Tools/Services Needed - Gmail: To trigger the workflow and label emails. - Google Gemini (or any LLM Model): To intelligently classify email content. How It Works - Gmail Trigger: Detects every new incoming email. - Text Classifier Node: Classifies the email content into predefined categories. - Google Gemini Chat Model: Provides the AI-powered understanding behind the classification. Conditional Labeling: - If the email is High Priority, label it accordingly. - If it’s Work Related (e.g., internal emails), apply the work label. - If it’s a Promotion, sort it into the promotions label. Gmail Labeling: Automatically adds the correct label to the email. Setup Instructions 1. Connect your Gmail account to n8n. 2. Connect your Google Gemini (or other LLM) credentials. 3. Customize the categories and labels if needed. 4. Activate the workflow — and that's it! Notes You can easily add more categories (like Finance, Newsletters, etc.) by adjusting the classification prompt. Works best with a clean and minimal set of categories to avoid overlap. Can be adapted to work with any other large language model (OpenAI, Claude, etc.) if preferred.

AI YouTube Playlist & Video Analyst Chatbot This n8n workflow transforms entire YouTube playlists or single videos into interactive knowledge bases you can chat with. Ask questions and get summaries without needing to watch hours of content. 🌟 How it Works🔗 Provide a Link: Start by giving the workflow a URL for a YouTube playlist or a single video. 📄 Content Retrieval: The workflow automatically fetches the video details and transcripts for the provided link. For playlists, it can process multiple videos at once (you might be asked how many). 🧠 AI Processing: Google's Gemini AI reads through the transcripts, understands the content, and creates summaries. 💾 Storage & Context: The processed information and summaries are stored in a vector database (Qdrant), making them ready for conversation. Context is managed using Redis, remembering the current video/playlist you're discussing. 💬 Chat & Ask: Now, you can ask the AI agent questions about the playlist or video content! Because context is maintained, you can ask follow-up questions (like "expand on point X") without needing to provide the URL again. 🛠️ Requirements - Community Node: This workflow uses the youtubeTranscripter community node to fetch video transcripts. You'll need to install it in your n8n environment. - Installation: npm install n8n-nodes-youtube-transcription-dmr - Important: Community nodes require a self-hosted n8n instance. - Redis: A Redis instance is required for managing conversation context and status between interactions. - Credentials: You will need API credentials configured in your n8n instance for: - Google Gemini (AI Models) - Qdrant (Vector Store) - Redis (Context Store) 🤖 AI Agent Capabilities Engage with the AI agent to explore the video content. Since the agent remembers the context of your conversation, you can ask detailed follow-up questions naturally: - Get a quick summary of a single video or an entire playlist. - Ask for key takeaways or main topics discussed. - Query for specific information mentioned in the videos. - Ask the agent to elaborate on a specific point previously mentioned. - Understand complex subjects without watching the full duration. 🚀 Use Cases - 📊 Content Analysis: Quickly understand the themes and key points across a playlist or long video. - 📚 Research & Learning: Extract insights from educational series or tutorials efficiently. - ✍️ Content Creation: Easily repurpose video transcript information into blog posts, notes, or social media content. - ⏱️ Save Time: Get the essence of video content when you're short on time. - ♿ Accessibility: Offers a text-based way to interact with and understand video content. ✨ Sample Prompts - Please analyze this playlist and tell me the main topics covered: [YouTube Playlist URL] - Summarize the first 5 videos in this playlist: [YouTube Playlist URL] - (Follow-up) Tell me more about the main point in video 3. - What are the key points discussed in this video? [YouTube Video URL] - (Follow-up) Expand on the second key point you mentioned. - Does the video at [YouTube Video URL] mention [specific topic]?

Seamlessly automate the bulk creation of images using Leonardo.AI based on prompts from a Google Sheets list. Once generated, the images are automatically saved to a specified Google Drive folder for easy access and organization.

Use Case It is very convenient to add expenses via a simple chat message. This workflow attempts to do exactly this using AI-powered n8n magic! Send a message to a chat, something like "car wash; 59.3 USD; 25 Jan 2024" and get a response: Your expense saved, here is the output of save sub-workflow: {"cost":59.3,"descr":"car wash","date":"2024-01-25","msg":"car wash; 59.3 usd; 25 jan 2024"} LLM will smartly parse your message to structured JSON and save the expense as a new row into Google Sheets! Installation 1. Set up Google Sheets: Clone this Sheet: (File -> Make a copy) Choose this sheet into "Save expense into Google Sheets" node. 2. Fix sub-workflow dropdown: Open "Parse msg and save to Sheets" node (which is an n8n sub-workflow executor tool) and make sure the SAME workflow is chosen in the dropdown. It will allow n8n to locate and call "Workflow Input Trigger" properly when needed. 3. Activate the workflow to make chat work properly. Send a message to chat, something like "car wash; 59.3 USD; 25 Jan 2024", and you should get a response:Your expense saved, here is the output of save sub-workflow: {"cost":59.3,"descr":"car wash","date":"2024-01-25","msg":"car wash; 59.3 usd; 25 jan 2024"} And a new row in Google Sheets should be inserted!

This n8n workflow is a proof-of-concept template exploring how we might work with multimodal LLMs and their multi-image analysis capabilities. In this demo, we compare two screenshots of a webpage taken at different timestamps and pass both to our multimodal LLM for a visual comparison of differences. Handling multiple binary inputs (i.e., images) in an AI request is supported by n8n's basic LLM node. How it works This template is intended to run in two parts: first to generate the base screenshots and next to run the visual regression test which captures fresh screenshots. Starting with a list of webpages captured in a Google Sheet, base screenshots are captured for each using an external web scraping service called Apify.com (I prefer Apify, but feel free to use whichever web scraping service available to you). These base screenshots are uploaded to Google Drive and will be referenced later when we run our testing. In phase two of the workflow, we'll use a scheduled trigger to fire sometime in the future which will reuse our web scraping service to generate fresh screenshots of our desired webpages. Next, we re-download our base screenshots in parallel and with both old and new captures, we'll pass these to our LLM node. In the LLM node's options, we'll define two "user message" inputs with the type of binary (data) for our images. Finally, we'll prompt our LLM with our testing criteria and capture the regressions detected. Note, results will vary depending on which LLM you use. A final report can be generated using the LLM's output and is uploaded to Linear. Requirements - Apify.com API key for web screenshotting service - Google Drive and Sheets access to store the list of webpages and capturesCustomizing this workflow Have your own preferred web screenshotting service? Feel free to swap out Apify with your service of choice. If the web screenshot is too large, it may prove difficult for the LLM to spot differences with precision. Try splitting up captures into smaller images instead.

Automatically turn YouTube videos into clear, structured content ideas stored in Airtable. This workflow pulls new video links from Airtable, extracts transcripts using a RapidAPI service, summarizes them with your favourite LLM, and logs the main idea and key takeaways—keeping your content pipeline fresh with minimal effort. What It Does - Scans Airtable for new YouTube video links every 5 minutes. - Extracts the transcript of the video using a third-party API via RapidAPI. - Summarizes the content to generate a main idea and takeaways. - Updates the original Airtable entry with the insights and marks it as completed. Prerequisites Before using this template, make sure you have: - A RapidAPI account with access to the youtube-video-summarizer-gpt-ai API. - A valid RapidAPI key. - An OpenAI, Claude or Gemini account connected to n8n. - An Airtable account with a base and table ready. Setup Instructions 1. Clone this template into your n8n workspace. 2. Open the Get YouTube Sources node and configure your Airtable credentials. 3. In the Get video transcript node: - Enter your X-RapidAPI-Key under headers. - The API endpoint is pre-configured. 4. Connect your LLM credentials to the Extract detailed summary node. 5. (Optional) Adjust the summarization prompt in the LangChain node to better suit your tone. 6. Set your preferred schedule in the Trigger node. Airtable Setup Create a base (e.g., Content Hub) with a table named Ideas and the following columns: | Column Name | Type | Required | Notes | |-------------|----------------|----------|-------------------------------------------| | Type | Single select | ✅ | Must be set to YouTube Video | | Source | URL | ✅ | The YouTube video URL | | Status | Checkbox | ✅ | Leave empty initially; updated after processing | | Main Idea | Single line text | ✅ | Summary generated by OpenAI | | Key Takeaways | Long text | ✅ | List of takeaways extracted from the transcript | Activate the workflow—and you're done!