by Belgacem Dhiflaoui
What Problem Does This Solve? This workflow automates the end-to-end process of capturing company information from Google Drive, storing it semantically in Pinecone, and interacting with users via an intelligent AI chatbot. It eliminates the need for manual customer service, lead tracking, and company information retrieval—offering a fully automated, intelligent engagement system. Perfect for teams that need to: Maintain accurate, AI-readable company knowledge bases Answer customer inquiries 24/7 using AI Automatically collect and log lead information Embed a chatbot into their website to assist potential customers Target Audience: Sales teams, business owners, marketing departments, customer support reps, startup founders, or anyone looking to automate AI-powered lead generation and customer engagement. What Does It Do? Part One – Knowledge Ingestion Monitors** a Google Drive folder for new .txt or document uploads. Downloads** the document and splits the content into manageable chunks using a recursive character splitter. Generates** embeddings via OpenAI. Stores** the embeddings in a Pinecone vector database under the Q&A namespace. Purpose:** This knowledge base is later used to answer business-related questions through AI. Part Two – AI Chatbot Engagement Listens** for incoming chat messages using n8n’s chatTrigger node. Activates an AI agent** (powered by GPT-4o) to respond to inquiries regarding business hours, services, products, or general company info. Retrieves knowledge** using a vector search tool connected to Pinecone (newCompany_q). Captures leads:** If a user shows interest, the AI collects and stores: Name Email Phone number Specific interest into a connected Google Sheet automatically. Key Features 🔄 Google Drive integration for real-time file processing 🧠 OpenAI embedding + Pinecone vector store for semantic memory 🤖 LangChain agent with tool-based reasoning 🗃️ Google Sheets integration for dynamic lead storage 💬 GPT-4o model for accurate, human-like conversation ⚙️ Modular design to expand into CRM, Notion, or email workflows 🌐 Website-ready chatbot endpoint 🧰 Setup Instructions Prerequisites: n8n instance (cloud or self-hosted) Google Drive account (for uploading company data) Pinecone account (for vector storage) OpenAI API key Google Sheets access with OAuth2 credentials 📦 Installation Steps 1. Import the Workflow Upload the JSON files into your n8n instance. 2. Configure Credentials In n8n > Credentials, connect: Google Drive OpenAI Pinecone Google Sheets **3. Set Pinecone Index & Namespace Example:** Index: comanyName Namespace: Q&A 4. Test the Flow Upload a sample .txt or pdf file to the monitored Drive folder. Send a message to the chatbot (e.g., "What are your opening hours?"). Check the Google Sheet for collected user info. How It Works (Behind the Scenes) Part 1 – Data Preparation: Company files are uploaded to Google Drive. File is detected, downloaded, and chunked. Embeddings are created using OpenAI. Data is stored in Pinecone for semantic retrieval. Part 2 – Chat Interaction: A chat message triggers the workflow via webhook. The AI agent interprets the intent and accesses company data via newCompany_q. If lead data is gathered, it is appended to a Google Sheet using the AI-parsed values. Need help customizing? Contact me for consulting and support or add me on Linkedin.
by Yaron Been
Automatically monitor and track funding rounds in the US Fintech and Healthtech sectors using Crunchbase API, with daily updates pushed to Google Sheets for easy analysis and monitoring. 🚀 What It Does Daily Monitoring**: Automatically checks for new funding rounds every day at 8 AM Smart Filtering**: Focuses on US-based Fintech and Healthtech companies Data Enrichment**: Extracts and formats key funding information Automated Storage**: Pushes data to Google Sheets for easy access and analysis 🎯 Perfect For VC firms tracking investment opportunities Startup founders monitoring market activity Market researchers analyzing funding trends Business analysts tracking competitor funding ⚙️ Key Benefits ✅ Real-time funding round monitoring ✅ Focused industry tracking (Fintech & Healthtech) ✅ Automated data collection and organization ✅ Structured data output in Google Sheets ✅ Complete funding details including investors and amounts 🔧 What You Need Crunchbase API key Google Sheets account n8n instance Basic spreadsheet setup 📊 Data Collected Company Name Industry Funding Round Type Announced Date Money Raised (USD) Investors Crunchbase URL 🛠️ Setup & Support Quick Setup Deploy in 30 minutes with our step-by-step configuration guide 📺 Watch Tutorial 💼 Get Expert Support 📧 Direct Help Stay ahead of market movements with automated funding round tracking. Transform manual research into an efficient, automated process.
by Jimleuk
This n8n template shows you how to create an MCP server out of your existing n8n workflows. With this, any MCP client connected can get more done with powerful end-to-end workflows rather than just simple tools. Designing agent tools for outcome rather than utility has been a long recommended practice of mine and it applies well when it comes to building MCP servers; In gist, agents to be making the least amount of calls possible to complete a task. This is why n8n can be a great fit for MCP servers! This template connects your agent/MCP client (like Claude Desktop) to your existing workflows by allowing the AI to discover, manage and run these workflows indirectly. How it works An MCP trigger is used and attaches 4 custom workflow tools to discover and manage existing workflows to use and 1 custom workflow tool to execute them. We'll introduce an idea of "available" workflows which the agent is allowed to use. This will help limit and avoid some issues when trying to use every workflow such as clashes or non-production. The n8n node is a core node which taps into your n8n instance API and is able to retrieve all workflows or filter by tag. For our example, we've tagged the workflows we want to use with "mcp" and these are exposed through the tool "search workflows". Redis is used as our main memory for keeping track of which workflows are "available". The tools we have are "add Workflow", "remove workflow" and "list workflows". The agent should be able to manage this autonomously. Our approach to allow the agent to execute workflows is to use the Subworkflow trigger. The tricky part is figuring out the input schema for each but was eventually solved by pulling this information out of the workflow's template JSON and adding it as part of the "available" workflow's description. To pass parameters through the Subworkflow trigger, we can do so via the passthrough method - which is that incoming data is used when parameters are not explicitly set within the node. When running, the agent will not see the "available" workflows immediately but will need to discover them via "list" and "search". The human will need to make the agent aware that these workflows will be preferred when answering queries or completing tasks. How to use First, decide which workflows will be made visible to the MCP server. This example uses the tag of "mcp" but you can all workflows or filter in other ways. Next, ensure these workflows have Subworkflow triggers with input schema set. This is how the MCP server will run them. Set the MCP server to "active" which turns on production mode and makes available to production URL. Use this production URL in your MCP client. For Claude Desktop, see the instructions here - https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-langchain.mcptrigger/#integrating-with-claude-desktop. There is a small learning curve which will shape how you communicate with this MCP server so be patient and test. The MCP server will work better if there is a focused goal in mind ie. Research and report, rather than just a collection of unrelated tools. Requirements N8N API key to filter for selected workflows. N8N workflows with Subworkflow triggers! Redis for memory and tracking the "available" workflows. MCP Client or Agent for usage such as Claude Desktop - https://claude.ai/download Customising this workflow If your targeted workflows do not use the subworkflow trigger, it is possible to amend the executeTool to use HTTP requests for webhooks. Managing available workflows helps if you have many workflows where some may be too similar for the agent. If this isn't a problem for you however, feel free to remove the concept of "available" and let the agent discover and use all workflows!
by David Ashby
Complete MCP server exposing all Gong Tool operations to AI agents. Zero configuration needed - all 4 operations pre-built. ⚡ Quick Setup Need help? Want access to more workflows and even live Q&A sessions with a top verified n8n creator.. All 100% free? Join the community Import this workflow into your n8n instance Activate the workflow to start your MCP server Copy the webhook URL from the MCP trigger node Connect AI agents using the MCP URL 🔧 How it Works • MCP Trigger: Serves as your server endpoint for AI agent requests • Tool Nodes: Pre-configured for every Gong Tool operation • AI Expressions: Automatically populate parameters via $fromAI() placeholders • Native Integration: Uses official n8n Gong Tool tool with full error handling 📋 Available Operations (4 total) Every possible Gong Tool operation is included: 🔧 Call (2 operations) • Get call • Get many calls 👤 User (2 operations) • Get user • Get many users 🤖 AI Integration Parameter Handling: AI agents automatically provide values for: • Resource IDs and identifiers • Search queries and filters • Content and data payloads • Configuration options Response Format: Native Gong Tool API responses with full data structure Error Handling: Built-in n8n error management and retry logic 💡 Usage Examples Connect this MCP server to any AI agent or workflow: • Claude Desktop: Add MCP server URL to configuration • Custom AI Apps: Use MCP URL as tool endpoint • Other n8n Workflows: Call MCP tools from any workflow • API Integration: Direct HTTP calls to MCP endpoints ✨ Benefits • Complete Coverage: Every Gong Tool operation available • Zero Setup: No parameter mapping or configuration needed • AI-Ready: Built-in $fromAI() expressions for all parameters • Production Ready: Native n8n error handling and logging • Extensible: Easily modify or add custom logic > 🆓 Free for community use! Ready to deploy in under 2 minutes.
by David Ashby
Complete MCP server exposing all Jina AI Tool operations to AI agents. Zero configuration needed - all 3 operations pre-built. ⚡ Quick Setup Need help? Want access to more workflows and even live Q&A sessions with a top verified n8n creator.. All 100% free? Join the community Import this workflow into your n8n instance Activate the workflow to start your MCP server Copy the webhook URL from the MCP trigger node Connect AI agents using the MCP URL 🔧 How it Works • MCP Trigger: Serves as your server endpoint for AI agent requests • Tool Nodes: Pre-configured for every Jina AI Tool operation • AI Expressions: Automatically populate parameters via $fromAI() placeholders • Native Integration: Uses official n8n Jina AI Tool tool with full error handling 📋 Available Operations (3 total) Every possible Jina AI Tool operation is included: 🔧 Reader (2 operations) • Read URL content • Search web 🔧 Research (1 operations) • Perform deep research 🤖 AI Integration Parameter Handling: AI agents automatically provide values for: • Resource IDs and identifiers • Search queries and filters • Content and data payloads • Configuration options Response Format: Native Jina AI Tool API responses with full data structure Error Handling: Built-in n8n error management and retry logic 💡 Usage Examples Connect this MCP server to any AI agent or workflow: • Claude Desktop: Add MCP server URL to configuration • Custom AI Apps: Use MCP URL as tool endpoint • Other n8n Workflows: Call MCP tools from any workflow • API Integration: Direct HTTP calls to MCP endpoints ✨ Benefits • Complete Coverage: Every Jina AI Tool operation available • Zero Setup: No parameter mapping or configuration needed • AI-Ready: Built-in $fromAI() expressions for all parameters • Production Ready: Native n8n error handling and logging • Extensible: Easily modify or add custom logic > 🆓 Free for community use! Ready to deploy in under 2 minutes.
by Saswat Saubhagya Rout
📝 Use Case This n8n workflow automates the creation and publication of technical blog posts based on a list of topics stored in Google Sheets. It fetches context using Tavily and Wikipedia, generates Markdown-formatted content with Gemini AI, commits it to a GitHub repository, and updates a Jekyll-powered blog — all without manual intervention. Ideal for developers, bloggers, or content teams who want to streamline technical content creation and publishing. ⚙️ Setup Instructions 🔑 Prerequisites n8n (cloud or self-hosted) Tavily API key Google Sheets with blog topics Gemini (Google Palm) API key GitHub repository (Jekyll enabled) GitHub OAuth2 credentials Google OAuth2 credentials 🧩 Setup Steps Import the workflow JSON into your n8n instance. Set up the following credentials in n8n: Tavily API Google Sheets OAuth2 Google Palm/Gemini AI GitHub OAuth2 Prepare your Google Sheet: Columns: Title, status, row_number Set status to blank for topics to be picked up. Configure: GitHub repo and _posts/ path Jekyll setup (front matter, _config.yml, GitHub Pages) Adjust prompt/custom parameters if needed. Enable and deploy the workflow. Schedule it daily or trigger manually. 🔄 Workflow Details | Node | Function | |------|----------| | Schedule Trigger | Triggers the flow at a set interval | | Google Sheets (Get Topic) | Fetches the next incomplete blog topic | | Extract Topic | Parses topic text from the sheet | | Tavily Search | Gathers up-to-date content related to the topic | | Wikipedia Tool | Optionally adds more context or images | | Summarize Results | Formats the context for the AI | | Gemini AI Agent (LangChain) | Generates a Markdown blog post with YAML front matter | | Set File Parameters | Prepares the filename, content, and commit message | | GitHub Commit | Uploads the .md file to the _posts/ directory | | Update Google Sheet | Marks topic as done after successful commit | 🛠️ Customization Options Change LLM prompt (e.g. tone, depth, format). Use OpenAI instead of Gemini by switching nodes. Modify filename pattern or GitHub repo path. Add Slack/Discord notifications after publish. Extend flow to upload images or embed YouTube links. ⚠️ Community Nodes Used This workflow uses the following community nodes: @tavily/n8n-nodes-tavily.tavily – for deep search > ⚠️ Ensure these are installed and enabled in your n8n instance. 💡 Pro Tips Use GitHub Actions to trigger an automatic Jekyll build post-commit. Structure blog posts with front matter, headings, and table of contents for SEO. Set Schedule Trigger to daily at a fixed time to keep content flowing. Enhance formatting in AI output using code blocks, images, and lists. ✅ Example Output title: "How LLMs Are Changing Web Development" date: "2025-07-25" categories: [webdev, AI] tags: [LLM, Gemini, n8n, automation] excerpt: "Learn how LLMs like Gemini are transforming how we generate and deploy developer content." author: "Saswat Saubhagya" Table of Contents Introduction Understanding LLMs Use Cases in Web Development Challenges Conclusion ...
by Joseph
Here is your refined template description with detailed step-by-step instructions, markdown formatting, and customization guidance. YouTube Transcript Extraction Workflow This n8n workflow extracts and processes transcripts from YouTube videos using the YouTube Transcript API on RapidAPI. It allows users to retrieve subtitles from YouTube videos, clean them up, and return structured transcript data for further processing. Table of Contents Problem Statement & Target Audience Pre-conditions & API Requirements Step-by-Step Workflow Explanation Customization Guide How to Set Up This Workflow Problem Statement & Target Audience Who is this for? This workflow is ideal for content creators, researchers, and developers who need to: Extract subtitles from YouTube videos automatically. Format and clean** transcript data for readability. Use transcripts for summarization, content repurposing, or language analysis. Pre-conditions & API Requirements API Required YouTube Transcript API** (RapidAPI) n8n Setup Prerequisites A running n8n instance (Installation Guide) A RapidAPI account to access the YouTube Transcript API An API key from RapidAPI to authenticate requests Step-by-Step Workflow Explanation 1. Input YouTube Video URL (Trigger) This step provides a simple input form where users enter a YouTube video URL. 2. HTTP Request Node (Retrieve Transcript Data) Makes a POST request to the YouTube Transcript API via RapidAPI. Passes the video URL received from the input form. Uses an environment variable to store the API key securely. 3. Function Node (Process Transcript) Receives* the API response containing the *raw transcript**. Processes and cleans** the transcript: Removes unwanted characters. Formats text for readability. Handles errors** when no transcript is available. Outputs* both the *raw and cleaned transcript** for further use. 4. Set Field Node (Response Formatting) Structures** the processed transcript data into a user-friendly format. Returns** the final transcript data to the client. Customization Guide 1. Modify Transcript Cleaning Rules Update the Function Node to apply custom text processing, such as: Removing timestamps. Changing the output format (e.g., JSON, plain text). 2. Store Transcripts in a Database Add a Database Node (e.g., MySQL, PostgreSQL, or Firebase) to save transcripts. 3. Generate Summaries from Transcripts Integrate AI services (e.g., OpenAI, Google Gemini) to summarize transcripts. 4. Convert Transcripts into Speech Use ElevenLabs API to generate an AI-powered voiceover from transcripts. How to Set Up This Workflow Step 1: Import the Workflow into n8n Download or copy the workflow JSON file. Import it into your n8n instance. Step 2: Set Up the API Key Sign up for the YouTube Transcript API. Subscribe to the api. Copy and paste your api key where the "your_api_key" is. Step 3: Activate the Workflow Start the workflow in n8n. Enter a YouTube video URL in the input form. The workflow will return a cleaned transcript. This workflow ensures seamless YouTube transcript extraction and processing with minimal manual effort. 🚀
by Ferenc Erb
Overview An automation workflow that creates a complete REST API for digitally signing PDF documents using n8n webhooks. This service demonstrates how to implement secure document signing functionality through standardized API endpoints with file upload and download capabilities. Use Case This workflow is designed for developers and automation specialists who need to implement digital document signing. It's particularly useful for: Integrating PDF signing capabilities into existing document workflows API-based automation of signature processes Creating proof-of-concept implementations for document verification systems Learning n8n's webhook capabilities and file handling techniques Testing PDF signing in development environments before production implementation What This Workflow Does API-Based Document Management Exposes RESTful webhook endpoints for all document operations Handles multipart/form-data uploads for PDF documents Processes JSON payloads for signing configuration Provides download functionality for completed documents Digital Certificate Handling Uploads existing PFX/PKCS#12 digital certificates Generates new certificates with customizable attributes Securely manages certificate storage and access Associates certificates with signing operations Cryptographic PDF Signing Applies digital signatures using industry-standard cryptographic methods Embeds signature information within PDF document structure Validates document integrity through cryptographic verification Preserves original document while adding signature elements Webhook Integration System Routes different API methods to appropriate handlers Validates request payloads and file content Manages authentication through webhook paths Returns structured responses for integration with other systems Technical Architecture Components API Gateway: n8n webhook nodes that receive external requests Request Router: Switch nodes that direct operations based on method parameters Document Processor: Function nodes for PDF manipulation and verification Certificate Manager: Specialized nodes for cryptographic key operations Storage Interface: File operation nodes for document persistence Response Formatter: Nodes that structure API responses Integration Flow Client Request → Webhook Endpoint → Method Router → Processing Engine → Digital Signing → Storage → Response Generation → Client Response Setup Instructions Prerequisites n8n installation (minimum version 0.214.0) Node.js 14 or higher Required environment variable: NODE_FUNCTION_ALLOW_EXTERNAL: "node-forge,@signpdf/signpdf,@signpdf/signer-p12,@signpdf/placeholder-plain" Configuration Steps Import Workflow Import the workflow JSON into your n8n instance Activate the workflow to enable the webhooks Configure Storage Set the storage path variables in the workflow Ensure proper permissions on the storage directories Test API Endpoints Use the included test scripts to verify functionality Test PDF upload, certificate generation, and signing Integration Document the webhook URLs for integration with other systems Configure error handling according to your requirements Testing Methods Test the workflow functionality using various HTTP requests and JSON data: Upload PDF documents to the document processing endpoint Upload or generate digital certificates Execute PDF signing operations Download signed documents from the download endpoint Webhook Endpoints The workflow exposes two primary webhook endpoints that form a complete API for PDF digital signing operations: 1. Document Processing Endpoint (/webhook/docu-digi-sign) This endpoint handles all document and certificate operations: Method: Upload PDF HTTP: POST Content-Type: multipart/form-data Parameters: method, uploadType, fileName, fileData Method: Upload Certificate HTTP: POST Content-Type: multipart/form-data Parameters: method, uploadType, fileName, fileData Method: Generate Certificate HTTP: POST Content-Type: application/json Parameters: method, subjectCN, issuerCN, serialNumber, validFrom, validTo, password Method: Sign PDF HTTP: POST Content-Type: application/json Parameters: method, inputPdf, pfxFile, pfxPassword 2. Document Download Endpoint (/webhook/docu-download) This endpoint handles the retrieval of processed documents: Method: Download Signed PDF HTTP: GET Content-Type: application/json Parameters: method, fileType, fileName Key Workflow Sections The workflow is organized into logical sections with clear responsibilities: Request Processing**: Parses incoming webhook data Method Routing**: Directs requests to appropriate handlers Document Management**: Handles file operations and storage Cryptographic Operations**: Manages signing and certificate functions Response Formatting**: Structures and returns results
by Akash Kankariya
🚀 Discover trending and viral YouTube videos easily with this powerful n8n automation! This workflow helps you perform bulk research on YouTube videos related to any search term, analyzing engagement data like views, likes, comments, and channel statistics — all in one streamlined process. ✨ Perfect for: Content creators wanting to find viral video ideas Marketers analyzing competitor content YouTubers optimizing their content strategy How It Works 🎯 1️⃣ Input Your Search Term — Simply enter any keyword or topic you want to research. 2️⃣ Select Video Format — Choose between short, medium, or long videos. 3️⃣ Choose Number of Videos — Define how many videos to analyze in bulk. 4️⃣ Automatic Data Fetch — The workflow grabs video IDs, then fetches detailed video data and channel statistics from the YouTube API. 5️⃣ Performance Scoring — Videos are scored based on engagement rates with easy-to-understand labels like 🚀 HOLY HELL (viral) or 💀 Dead. 6️⃣ Export to Google Sheets — All data, including thumbnails and video URLs, is appended to your Google Sheet for comprehensive review and easy sharing. Setup Instructions 🛠️ Google API Key Get your YouTube Data API key from Google Developers Console. Add it securely in the n8n credentials manager (do not hardcode). Google Sheets Setup Create a Google Sheet to store your results (template link is provided). Share the sheet with your Google account used in n8n. Update the workflow with your sheet's Document ID and Sheet Name if needed. Run the Workflow Trigger the form webhook via browser or POST call. Enter search term, format, and number of videos. Let it process and check your Google Sheet for insights! Features ✨ Bulk fetches the latest and top-viewed YouTube videos. Intelligent video performance scoring with emojis for quick insights 🔥🎬. Organizes data into Google Sheets with thumbnail previews 🖼️. Easy to customize search parameters via an intuitive form. Fully automated, no manual API calls needed. Get Started Today! 🌟 Boost your YouTube content strategy and stay ahead with this powerful viral video research automation! Try it now on your n8n instance and tap into the world of viral content like a pro 🎥💡
by Ranjan Dailata
Notice Community nodes can only be installed on self-hosted instances of n8n. Who this is for Recipe Recommendation Engine with Bright Data MCP & OpenAI is a powerful automated workflow combines Bright Data's MCP for scraping trending or regional recipe data with OpenAI 4o mini to generate personalized recipe recommendations. This automated workflow is designed for: Food Bloggers & Culinary Creators : Who want to automate the extraction and curation of recipes from across the web to generate content, compile cookbooks, or publish newsletters. Nutritionists & Health Coaches : Who need structured recipe data to analyze ingredients, calories, and nutrition for personalized meal planning or dietary tracking. AI/ML Engineers & Data Scientists : Building models that classify cuisines, predict recipes from ingredients, or generate dynamic meal suggestions using clean, structured datasets. Grocery & Meal Kit Platforms : Who aim to extract recipes to power recommendation engines, ingredient lists, or personalized meal plans. Recipe Aggregator Startups : Looking to scale recipe data collection, filtering, and standardization across diverse cooking websites with minimal human intervention. Developers Integrating Cooking Features : Into apps or digital assistants that offer recipe recommendations, step-by-step cooking instructions, or nutritional insights. What problem is this workflow solving? This workflow solves: Automated recipe data extraction from any public URL AI-driven structured data extraction Scalable looped crawling and processing Real-time notifications and data persistence What this workflow does 1. Set Recipe Extract URL Configure the recipe website URL in the input node Set your Bright Data zone name and authentication 2. Paginated Data Extract Triggers a paginated extraction across multiple pages (recipe listing, index, or search pages) Returns a list of recipe links for processing 3. Loop Over Items Loops through the array of recipe links Each link is passed individually to the scraping engine 4. Bright Data MCP Client (Per Recipe) Scrapes each individual recipe page using scrape_as_html Smartly bypasses common anti-bot protections via Bright Data Web Unlocker 5. Structured Recipe Data Extract (via OpenAI GPT-4o mini) Converts raw HTML to clean text using an LLM preprocessing node Uses OpenAI GPT-4o mini to extract structured data 6. Webhook Notification Pushes the structured recipe data to your configured webhook endpoint Format: JSON payload, ideal for Slack, internal APIs, or dashboards 7. Save Response to Disk Saves the structured recipe JSON information to local file system Pre-conditions You need to have a Bright Data account and do the necessary setup as mentioned in the "Setup" section below. You need to have an OpenAI Account. Setup Sign up at Bright Data. Navigate to Proxies & Scraping and create a new Web Unlocker zone by selecting Web Unlocker API under Scraping Solutions. In n8n, configure the Header Auth account under Credentials (Generic Auth Type: Header Authentication). The Value field should be set with the Bearer XXXXXXXXXXXXXX. The XXXXXXXXXXXXXX should be replaced by the Web Unlocker Token. In n8n, configure the OpenAi account credentials. Make sure to set the fields as part of Set the Recipe Extract URL. Remember to set the webhook_url to send a webhook notification of recipe response. Set the desired local path in the Write the structured content to disk node to save the recipe response. How to customize this workflow to your needs You can tailor the Recipe Recommendation Engine workflow to better fit your specific use case by modifying the following key components: 1. Input Fields Node Update the Recipe URL to target specific cuisine sites or recipe types (e.g., vegan, keto, regional dishes). 2. LLM Configuration Swap out the OpenAI GPT-4o mini model with another provider (like Google Gemini) if you prefer. Modify the structured data prompt to extract custom fields that you wish. 3. Webhook Notification Configure the Webhook Notification node to point to your preferred integration (e.g., Slack, Discord, internal APIs). 4. Storage Destination Change the Save to Disk node to store the structured recipe data in: A cloud bucket (S3, GCS, Azure Blob etc.) A database (MongoDB, PostgreSQL, Firestore) Google Sheets or Airtable for spreadsheet-style access.
by Eduard
This workflow demonstrates three distinct approaches to chaining LLM operations using Claude 3.7 Sonnet. Connect to any section to experience the differences in implementation, performance, and capabilities. What you'll find: 1️⃣ Naive Sequential Chaining The simplest but least efficient approach - connecting LLM nodes in a direct sequence. Easy to set up for beginners but becomes unwieldy and slow as your chain grows. 2️⃣ Agent-Based Processing with Memory Process a list of instructions through a single AI Agent that maintains conversation history. This structured approach provides better context management while keeping your workflow organized. 3️⃣ Parallel Processing for Maximum Speed Split your prompts and process them simultaneously for much faster results. Ideal when you need to run multiple independent tasks without shared context. Setup Instructions: API Credentials: Configure your Anthropic API key in the credentials manager. This workflow uses Claude 3.7 Sonnet, but you can modify the model in each Anthropic Chat Model node, or pick an entirely different LLM. For Cloud Users: If using the parallel processing method (section 3), replace {{ $env.WEBHOOK_URL }} in the "LLM steps - parallel" HTTP Request node with your n8n instance URL. Test Data: The workflow fetches content from the n8n blog by default. You can modify this part to use a different content or a data source. Customization: Each section contains a set of example prompts. Modify the "Initial prompts" nodes to change the questions asked to the LLM. Compare these methods to understand the trade-offs between simplicity, speed, and context management in your AI workflows! Follow me on LinkedIn for more tips on AI automation and n8n workflows!
by David Levesque
Here's the corrected English text: Dropbox Folder Monitoring Workflow As we don't have (yet?) a Dropbox node "Watching new files" or "Watching folder", I created this central workflow to do it. How it works Triggered by Dropbox webhook I respond immediately to Dropbox to avoid webhook disabling Then I add/duplicate one branch per monitored folder, according to my needs In my case, I need to monitor several folders, like "vocal notes to process", "transcriptions to LinkedIn posts" or "quotes to add". This workflow shows 2 types of folder monitoring: Way #1: Each file in the monitored folder calls a sub-workflow Way #2: We get all files from the monitored folder and compare them to a database. If the file is not listed in DB, i supposed it's new one. Way #1 - We get all files from the monitored folder I set a variable folder_to_watch to indicate which folder to monitor. This step is here just to be homogeneous and allow setting the folder path only once in this branch. I list the folder files We keep only files (exclude folders) Then I call the specialized sub-workflow Way #2 - We want only new files from the monitored folder I set a variable folder_to_watch to indicate which folder to monitor I list the folder files and keep only files Meanwhile, I query my DB to get known files about this folder (I send the query to NocoDB (folder_to_watch,eq,{{ $json.folder_to_watch }})) Now I can exclude old files and keep only new ones by merging (I compare from Dropbox file id - as the file could be renamed by the user) I add the new file in DB to be sure to recognize it next time - I save the JSON Dropbox data: { "id":"{{ $json.id }}", "name":"{{ $json.name }}", "lastModifiedClient": "{{ $json.lastModifiedClient }}", "lastModifiedServer": "{{ $json.lastModifiedServer }}", "rev": "{{ $json.rev }}", "contentSize": {{ $json.contentSize }}, "type": "{{ $json.type }}", "contentHash": "{{ $json.contentHash }}", "pathLower": "{{ $json.pathLower }}", "pathDisplay": "{{ $json.pathDisplay }}", "isDownloadable": {{ $json.isDownloadable }} } And now I can call my sub-workflow :) My DB Columns details: folder_to_watch data (json/text) timestamp file_id (Dropbox file ID, to ease future searches) My vision: I have only one workflow in my n8n that monitors Dropbox folders/files This workflow calls the required sub-workflow specialized for the tasks required I will have as many branches as I have folders to monitor (if I have 5 different folders to watch, I will get 5 branches and 5 sub-workflows)