Workflow Templates

🤖 Human-like Evolution API Agent with Redis & PostgreSQL This production-ready template builds a sophisticated AI Agent using Evolution API that mimics human interaction patterns. Unlike standard chatbots that reply instantly to every incoming message, this workflow uses a Smart Redis Buffering System. It waits for the user to finish typing their full thought (text, audio, or image albums) before processing, creating a natural, conversational flow. It features a Hybrid Memory Architecture: active conversations are cached in Redis for ultra-low latency, while the complete chat history is securely stored in PostgreSQL. To optimize token usage and maintain long-term coherence, a Context Refiner Agent summarizes the conversation history before the Main AI generates a response. ✨ Key Features Human-like Buffering:** The agent waits (configurable time) to group consecutive messages, voice notes, and media albums into a single context. This prevents fragmented replies and feels like talking to a real person. Hybrid Memory:* Combines *Redis* (Hot Cache) for speed and *PostgreSQL** (Cold Storage) for permanent history. Context Refinement:** A specialized AI step summarizes past interactions, allowing the Main Agent to understand long conversations without exceeding token limits or increasing costs. Multi-Modal Support:** Natively handles text, audio transcription, and image analysis via Evolution API. Parallel Processing:** Manages "typing..." status and session checks in parallel to reduce response latency. 📋 Requirements To use this workflow, you must configure the Evolution API correctly: Evolution API Instance: You need a running instance of Evolution API. Configuration Guide N8n Community Node: Install the Evolution API node in your n8n instance. n8n-nodes-evolution-api Database: A PostgreSQL database for chat history and a Redis instance for the buffer/cache. AI Models: API keys for your LLM (OpenAI, Anthropic, or Google Gemini). ⚙️ Setup Instructions Install the Node: Go to Settings > Community Nodes in n8n and install n8n-nodes-evolution-api. Credentials: Configure credentials for Redis, PostgreSQL, and your AI provider (e.g., OpenAI/Gemini). Database Setup: Create a chat_history table in PostgreSQL (columns must match the Insert node). Redis Connection: Configure your Redis credentials in the workflow nodes. Global Variables: Set the following in the "Global Variables" node: wait_buffer: Seconds to wait for the user to stop typing (e.g., 5s). wait_conversation: Seconds to keep the cache alive (e.g., 300s). max_chat_history: Number of past messages to retrieve. Webhook: Point your Evolution API instance to this workflow's Webhook URL. 🚀 How it Works Ingestion: Receives data via Evolution API. Detects if it's text, audio, or an album. Smart Buffering: Holds the execution to collect all parts of the user's message (simulating a human reading/listening). Context Retrieval: Checks Redis for the active session. If empty, fetches from PostgreSQL. Refinement: The Refiner Agent summarizes the history to extract key details. Response: The Main Agent generates a reply based on the refined context and current buffer, then saves it to both Redis and Postgres. 💡 Need Assistance? If you’d like help customizing or extending this workflow, feel free to reach out: 📧 Email: johnsilva11031@gmail.com 🔗 LinkedIn: John Alejandro Silva Rodríguez

How It Works This workflow automates enterprise budget monitoring and cost optimization using Anthropic Claude as the core AI engine across multiple specialist agents. It targets finance teams, operations managers, and CFOs managing complex multi-department budgets where manual tracking leads to delayed decisions and cost overruns. The workflow triggers on schedule, generates metrics data, and routes it through a Cost Intelligence Agent that classifies budget status (Critical, Warning, Review, Feedback). Each path activates specialist agents—Budget Alert, Routing Recommendation, and Cost Projection—coordinated by an Optimization Coordinator. Results are routed by action type: urgent alerts fire via Slack, executive summaries deliver via email, and all optimization actions are stored. This gives finance teams real-time cost intelligence with automated escalation and audit-ready records. Setup Steps Import workflow JSON into your n8n instance. Add Anthropic API credentials. Set Schedule Trigger frequency. Update Workflow Configuration node with budget thresholds per department or cost centre. Add Slack credentials and configure the target channel in the Send Slack Alert node. Set Gmail/SMTP credentials for the Send Executive Report Email node. Prerequisites n8n (cloud or self-hosted), Anthropic API key (Claude), Slack workspace with bot token Use Cases Finance teams automating multi-department budget variance detection and escalation Customization Replace Anthropic Claude with OpenAI GPT-4 or NVIDIA NIM in any agent node Benefits Eliminates manual budget reviews through automated AI-driven cost classification

How It Works This workflow automates industrial asset health monitoring and predictive maintenance using Anthropic Claude across coordinated specialist agents. It targets facility managers, maintenance engineers, and operations teams in manufacturing, energy, and infrastructure sectors where reactive maintenance leads to costly unplanned downtime and asset failures. On schedule, the system ingests asset health data and routes it through a Performance Evaluation Agent that coordinates three specialist agents: Maintenance Scheduling, Parts Readiness, and Lifecycle Reporting. An MCP External Data Tool enriches analysis with real-time contextual data. Results are risk-routed—Critical assets trigger immediate Slack alerts, High-risk assets escalate via email reports, and Routine cases are logged for scheduled maintenance. All paths merge into a unified maintenance log, giving operations teams proactive, audit-ready asset intelligence before failures occur. Setup Steps Import workflow JSON into your n8n instance. Add Anthropic API credentials. Set Schedule Trigger frequency aligned to your asset monitoring cycle. Update Workflow Configuration node with asset thresholds. Configure MCP External Data Tool with your external data source endpoint and authentication. Add Slack credentials and set the target channel in the Notify Critical Alert node. Set Gmail/SMTP credentials for the Email Escalation Report node. Prerequisites n8n (cloud or self-hosted), Anthropic API key (Claude), Slack workspace with bot token Use Cases Facility managers automating condition-based maintenance scheduling across multiple assets Customization Replace Anthropic Claude with OpenAI GPT-4 or NVIDIA NIM in any agent node Benefits Shifts maintenance from reactive to predictive, reducing unplanned downtime significantly

How It Works This workflow automates end-to-end medical claims processing using a multi-agent AI orchestration system built on OpenAI GPT-4. It targets healthcare revenue cycle teams, billing departments, and hospital administrators burdened by manual claims adjudication, coding errors, and payer denials. The workflow triggers on a schedule, loads billing data, and routes it through an Orchestrator Agent that coordinates four specialist sub-agents: Coding Validation, Claims Submission, Denial Detection, and Payer Follow-up. Each agent independently validates, submits, or flags claims. Results are parsed, merged, and routed by risk level. Final metrics and a formatted report close the cycle, giving teams real-time visibility into claim status, denial patterns, and revenue recovery. Setup Steps Import workflow JSON into your n8n instance. Add OpenAI API credentials. Configure Schedule Trigger with desired processing frequency. Update Workflow Configuration node with your billing system endpoint or sample data path. Set Gmail/SMTP credentials for the Escalate to Revenue Specialist email node. Connect Google Sheets or database nodes with appropriate credentials and sheet IDs. Test with simulated billing data before enabling live data sources. Prerequisites n8n, OpenAI API key (GPT-4) and Gmail or SMTP account Use Cases Hospital billing departments automating claims submission and denial follow-up Customization Swap OpenAI for NVIDIA NIM or Anthropic models in any agent node and add Slack alerts alongside email escalation Benefits Reduces manual claims review by 80%+ through parallel AI agent processing

⚖️ HR Sovereign: AI-Powered Onboarding Hub A high-fidelity employee onboarding engine: Intake → Role-Based Enrichment → AI Personalization → IT Provisioning. ⚙️ Core Sovereign Logic Enrichment:** Auto-classifies Tech, Sales, and Leadership roles to drive specific logic tracks. Intelligence:* Uses *AI Agent (GPT-4)** to generate personalized welcome messaging based on job DNA. Atomization:* *Merge PDF** node assembles role-specific policies and benefits into a single high-res package. Provisioning:* Dynamically generates *Jira* hardware/access tickets and *Notion** tracking dashboards. Delivery:* Sends branded HTML emails via *Gmail* and announces hires on *Slack**. 📋 Setup & Prerequisites Intake: Connect your HRIS (BambooHR/Workday) to the Webhook URL. Assets: Organize Drive folders into "Technical", "Leadership", and "Standard" templates. Tracking: Connect your Notion Onboarding Database and Jira IT Project. Metrics: Time_to_Provision, Engagement_Score, Document_Integrity_Hash.

AI Meeting Task Manager - Google Meet to GoHighLevel CRM 📋 TEMPLATE DESCRIPTION Transform your meeting follow-ups from chaos to clarity! This workflow automates the entire post-meeting workflow by scanning Google Meet recordings folders, extracting action items from AI-generated meeting notes, and creating tasks directly in your GoHighLevel CRM. 🎯 Who is this for? Sales teams using GoHighLevel CRM Agency owners managing multiple client meetings Anyone who uses Google Meet with Gemini note-taking Professionals drowning in meeting follow-ups ✨ What it does: Daily File Organization Scans your Google Meet recordings folder Automatically sorts recordings, notes, and chat logs into organized subfolders Keeps your Drive clean and searchable AI-Powered Task Extraction Reads Google Docs meeting notes (generated by Gemini) Identifies action items assigned to you Intelligently determines due dates from context (defaults to 3 business days) CRM Integration Searches for meeting participants in GoHighLevel Creates properly formatted tasks with full context Links tasks to the correct contact record Beautiful Email Summaries Sends a professionally designed HTML email Shows tasks created per contact Includes due dates and status updates 🔧 Technologies Used: Google Drive API (file management) Google Docs API (content extraction) GoHighLevel API (contact search + task creation) OpenAI GPT-4 (task extraction intelligence) Gmail API (email delivery) ⚙️ Setup Requirements: Google Cloud OAuth credentials (Drive, Docs, Gmail) GoHighLevel OAuth credentials OpenAI API key Create 4 Google Drive folders (source + 3 destination folders) 📖 Setup Instructions: Create Google Drive Folders: Source folder: Where Google Meet saves recordings Recordings folder: For video files Notes folder: For Gemini notes Chat folder: For meeting chat logs Configure Credentials: Connect Google Drive OAuth Connect Google Docs OAuth Connect GoHighLevel OAuth Connect Gmail OAuth Add OpenAI API key Update Folder URLs: Replace placeholder URLs in Google Drive nodes with your folder URLs Customize: Set your email address in the Gmail tool Set your GoHighLevel user ID for task assignment Adjust the schedule trigger timing as needed 💡 Pro Tips: Works best with Google Meet's Gemini note-taking feature Customize the AI prompts to match your task naming conventions The HTML email template is fully customizable

How it works This workflow creates a sophisticated, self-improving customer support system that automatically handles incoming emails. It's designed to answer common questions using an AI-powered knowledge base and, crucially, to learn from human experts when new or complex questions arise, continuously expanding its capabilities. Think of it like having an AI assistant with a smart memory and a human mentor. Here's the step-by-step process: New Email Received: The workflow is triggered whenever a new email arrives in your designated support inbox (via Gmail). Classify Request: An AI model (Google Gemini 2.5 Flash Lite) first classifies the incoming email to ensure it's a genuine support request, filtering out irrelevant messages. Retrieve Knowledge Base: The workflow fetches all existing Question and Answer pairs from your dedicated Google Sheet knowledge base. AI Answer Attempt: A powerful AI model (Google Gemini 2.5 Pro) analyzes the customer's email against the entire knowledge base. It attempts to find a highly relevant answer and drafts a complete HTML email response if successful. Decision Point: An IF node checks if the AI found a confident answer. If Answer Found: The AI-generated HTML response is immediately sent back to the customer via Gmail. If No Answer Found (Human-in-the-Loop): Escalate to Human: The customer's summarized question and original email are forwarded to a human expert (you or your team) via Gmail, requesting their assistance. Human Reply & AI Learning: The workflow waits for the human expert's reply. Once received, another AI model (Google Gemini 2.5 Flash) processes both the original customer question and the expert's reply to distill them into a new, generic, and reusable Question/Answer pair. Update Knowledge Base: This newly created Q&A pair is then automatically added as a new row to your Google Sheet knowledge base, ensuring the system can answer similar questions automatically in the future. Set up steps Setup time: ~10-15 minutes This workflow requires connecting your Gmail and Google Sheets accounts, and obtaining a Google AI API key. Follow these steps carefully: Connect Your Gmail Account: Select the On New Email Received node. Click the Credential dropdown and select + Create New Credential to connect your Gmail account. Grant the necessary permissions. Repeat this for the Send AI Answer and Ask Human for Help nodes, selecting the credential you just created. Connect Your Google Sheets Account: Select the Get Knowledge Base node. Click the Credential dropdown and select + Create New Credential to connect your Google account. Grant the necessary permissions. Repeat this for the Add to Knowledge Base node, selecting the credential you just created. Set up Your Google Sheet Knowledge Base: Create a new Google Sheet in your Google Drive. Rename the first sheet (tab) to QA Database. In the first row of QA Database, add two column headers: Question (in cell A1) and Answer (in cell B1). Go back to the Get Knowledge Base node in n8n. In the Document ID field, select your newly created Google Sheet. Do the same for the Add to Knowledge Base node. Get Your Google AI API Key (for Gemini Models): Visit Google AI Studio at aistudio.google.com/app/apikey. Click "Create API key in new project" and copy the key. In the workflow, go to the Google Gemini 2.5 Pro node, click the Credential dropdown, and select + Create New Credential. Paste your key into the API Key field and Save. Repeat this for the Google Gemini 2.5 Flash Lite and Google Gemini 2.5 Flash nodes, selecting the credential you just created. Configure Human Expert Email: Select the Ask Human for Help node. In the Send To field, replace the placeholder email address with the actual email address of your human expert (e.g., your own email or a team support email). Activate the Workflow: Once all credentials and configurations are set, activate the workflow using the toggle switch at the top right of your n8n canvas. Start Learning! Send a test email to the Gmail account connected to the On New Email Received node. Observe how the AI responds, or how it escalates to your expert email and then learns from the reply. Check your Google Sheet to see new Q&A pairs being added!

This workflow automates the end-to-end analysis of WooCommerce product reviews, transforming raw customer feedback into actionable product and customer-care insights, and delivering them in a structured, visual, and shareable format. This workflow analyzes product review sentiment from WooCommerce using AI. It starts by retrieving reviews for a specified product via the WooCommerce. Each review then undergoes sentiment analysis using LangChain's Sentiment Analysis. The workflow aggregates sentiment data, creates a pie chart visualization via QuickChart, and compiles a comprehensive report using an AI Agent. The report includes executive summaries, quantitative data, qualitative analysis, product diagnostics, and operational recommendations. Finally, the AI-generated report is converted to HTML and emailed to a designated recipient for review by customer and product teams. Key Advantages 1. ✅ Full Automation of Review Analysis Eliminates manual work by automating data collection, sentiment analysis, reporting, visualization, and delivery in a single workflow. 2. ✅ Scalable and Reliable Batch processing ensures the workflow can handle dozens or hundreds of reviews without performance issues. 3. ✅ Action-Oriented Insights (Not Just Sentiment) Instead of stopping at sentiment scores, the workflow produces: Root-cause hypotheses Concrete improvement actions Prioritized recommendations (P0 / P1 / P2) Measurable KPIs 4. ✅ Combines Quantitative and Qualitative Analysis Merges hard metrics (averages, distributions, outliers) with qualitative insights (themes, risks, opportunities), giving a 360° view of customer feedback. 5. ✅ Visual + Narrative Output Stakeholders receive both: Visual sentiment charts** for quick understanding Structured written reports** for strategic decision-making 6. ✅ Ready for Product & Customer Care Teams The output format is tailored for non-technical teams: Clear language Masked personal data (GDPR-friendly) Immediate usability in meetings, emails, or documentation 7. ✅ Easily Extensible The workflow can be extended to: Run on a schedule Analyze multiple products Store results in a database or CRM Trigger alerts for negative sentiment spikes Ideal Use Cases Continuous monitoring of product sentiment Supporting product roadmap decisions Identifying customer pain points early Improving customer support response strategies Reporting customer voice to stakeholders automatically How it works Manual Trigger & Configuration The workflow starts manually and sets the target WooCommerce product ID and store URL. Data Retrieval from WooCommerce Fetches all reviews for the selected product via the WooCommerce REST API. Retrieves product details (name, description, categories) to enrich the analysis context. Batch Processing of Reviews Reviews are processed in batches to ensure scalability and reliability, even with a large number of reviews. AI-Powered Sentiment Analysis Each review is analyzed using an OpenAI-based sentiment analysis model. For every review, the workflow extracts: Sentiment category (Positive / Negative / Neutral) Strength (intensity) Confidence (reliability of the classification) Data Normalization & Aggregation Review text is cleaned and structured. Sentiment data is aggregated to compute overall distributions and metrics. Visual Sentiment Distribution A pie chart is dynamically generated via QuickChart to visually represent sentiment distribution. Advanced AI Insight Generation A specialized AI agent (“Product Insights Analyst”) transforms the raw and aggregated data into a professional, structured report, including: Executive summary Quantitative statistics Qualitative themes Product diagnosis Operational recommendations Product backlog ideas Next steps HTML Conversion & Delivery The report is converted into clean HTML. The final output is automatically sent via email to stakeholders (e.g. product or customer care teams). Set up steps Configure credentials: Set up WooCommerce API credentials in the HTTP Request node. Add OpenAI API credentials for both sentiment analysis and reporting. Configure Gmail OAuth2 credentials for sending the final email report. Set parameters: In the "Product ID" node, replace PRODUCT_ID and YOUR_WEBSITE with actual product ID and WooCommerce site URL. Update the recipient email address in the "Send a message" node. Optional adjustments: Modify the pie chart design in the "QuichChart" node if needed. Adjust the report structure or language in the "Product Insights Analyst" system prompt. Run the workflow: Click "Execute workflow" on the manual trigger to start the process. Monitor execution in n8n to ensure all nodes process correctly. Once configured, the workflow will automatically analyze product reviews, generate insights, and deliver a formatted report via email. 👉 Subscribe to my new YouTube channel. Here I’ll share videos and Shorts with practical tutorials and FREE templates for n8n. Need help customizing? Contact me for consulting and support or add me on Linkedin.

Automate Your Life: The Ultimate AI Assistant in Telegram (Powered by Google Gemini) Transform your Telegram messenger into a powerful, multi-modal personal or team assistant. This n8n workflow creates an intelligent agent that can understand text, voice, images, and documents, and take action by connecting to your favorite tools like Google Calendar, Gmail, Todoist, and more. At its core, a powerful Manager Agent, driven by Google Gemini, interprets your requests, orchestrates a team of specialized sub-agents, and delivers a coherent, final response, all while maintaining a persistent memory of your conversations. Key Features 🧠 Intelligent Automation: Uses Google Gemini as a central "Manager Agent" to understand complex requests and delegate tasks to the appropriate tool. 🗣️ Multi-Modal Input: Interact naturally by sending text, voice notes, photos, or documents directly into your Telegram chat. 🔌 Integrated Toolset: Comes pre-configured with agents to manage your memory, tasks, emails, calendar, research, and project sheets. 🗂️ Persistent Memory: Leverages Airtable as a knowledge base, allowing the assistant to save and recall personal details, company information, or past conversations for context-rich interactions. ⚙️ Smart Routing: Automatically detects the type of message you send and routes it through the correct processing pipeline (e.g., voice is transcribed, images are analyzed). 🔄 Conversational Context: Utilizes a window buffer to maintain short-term memory, ensuring follow-up questions and commands are understood within the current conversation. How It Works The Telegram Trigger node acts as the entry point, receiving all incoming messages (text, voice, photo, document). A Switch node intelligently routes the message based on its type: Voice**: The audio file is downloaded and transcribed into text using a voice-to-text service. Photo**: The image is downloaded, converted to a base64 string, and prepared for visual analysis. Document**: The file is routed to a document handler that extracts its text content for processing. Text**: The message is used as-is. A Merge node gathers the processed input into a unified prompt. The Manager Agent receives this prompt. It analyzes the user's intent and orchestrates one or more specialized agents/tools: memory_base (Airtable): For saving and retrieving information from your long-term knowledge base. todo_and_task_manager (Todoist): To create, assign, or check tasks. email_agent (Gmail): To compose, search, or send emails. calendar_agent (Google Calendar): To schedule events or check your agenda. research_agent (Wikipedia/Web Search): To look up information. project_management (Google Sheets): To provide updates on project trackers. After executing the required tasks, the Manager Agent formulates a final response and sends it back to you via the Telegram node. Setup Instructions Follow these steps to get your AI assistant up and running. Telegram Bot: Create a new bot using the BotFather in Telegram to get your Bot Token. In the n8n workflow, configure the Telegram Trigger node's webhook. Add your Bot Token to the credentials in all Telegram nodes. For proactive messages, replace the chatId placeholders with your personal Telegram Chat ID. Google Gemini AI: In the Google Gemini nodes, add your credentials by providing your Google Gemini API key. Airtable Knowledge Base: Set up an Airtable base to act as your assistant's long-term memory. In the memory_base nodes (Airtable nodes), configure the credentials and provide the Base ID and Table ID. Google Workspace APIs: Connect your Google account credentials for Gmail, Google Calendar, and Google Sheets. In the relevant nodes, specify the Document/Sheet IDs you want the assistant to manage. Connect Other Tools: Add your credentials for Todoist and any other integrated tool APIs. Configure Conversational Memory: This workflow is designed for multi-user support. Verify that the Session Key in the "Window Buffer Memory" nodes is correctly set to a unique user identifier from Telegram (e.g., {{ $json.chat.id }}). This ensures conversations from different users are kept separate. Review Schedule Triggers: Check any nodes designed to run on a schedule (e.g., "At a regular time"). Adjust their cron expressions, times, and timezone to fit your needs (e.g., for daily summaries). Test the Workflow: Activate the workflow. Send a text message to your bot (e.g., "Hello!"). Estimated Setup Time 30–60 minutes:** If you already have your API keys, account credentials, and service IDs (like Sheet IDs) ready. 2–3 hours:** For a complete, first-time setup, which includes creating API keys, setting up new spreadsheets or Airtable bases, and configuring detailed permissions.

AI Daily Email Summary - Auto-Delivered to Slack, WhatsApp & Docs (Gmail + Gemini) Summary Stop starting your day by drowning in unread emails. This workflow runs every morning and turns your Gmail inbox into a clean, prioritized briefing — automatically organized into what needs your attention, what meetings are coming up, and what you can read later. The summary lands wherever you already live: Google Docs, Gmail, Slack, or WhatsApp. Powered by Google Gemini 2.5 Flash, with a built-in privacy layer that scrubs passwords, OTPs, and API keys before they ever leave your n8n instance. Key features Prioritized daily briefing:** Gemini reads your last 24 hours of email and groups it into Action Required, Meetings & Events, and Important Updates — so you know exactly where to focus. Deliver anywhere:** Route your summary to Google Docs (as a daily log), Gmail (as a formatted report), Slack (as a DM), or WhatsApp — toggle each channel on or off from a single config node. Safe to use with AI:** A regex-based privacy layer scrubs sensitive patterns (OTPs, passwords, API keys, card numbers) inside n8n before anything reaches the AI model. Zero wasted API calls:** If your inbox is empty, the workflow skips the AI step entirely and sends a clean "no emails" status instead. Built-in error alerts:** If any step fails — expired credentials, API error — you get a diagnostic email automatically so nothing breaks silently. How it works Trigger: Runs on a daily schedule (fully configurable). Fetch & clean: Pulls primary emails from the last 24 hours and scrubs sensitive patterns before processing. Summarize: Gemini 2.5 Flash generates a formatted text summary, categorized by priority. Deliver: Your summary is routed to whichever channels you've toggled on in the config node. Monitor: A global error handler catches failures and emails you a crash report. Nodes used Gmail (Fetch, Send, Error Alert) Google Gemini (LangChain Native Node) Slack (Post Alert) WhatsApp Business Cloud API Google Docs (Create & Append) Schedule Trigger Error Trigger Code (PII Scrubbing & Formatting) Setup instructions Credentials: Connect Gmail, Google Docs, Slack, WhatsApp, and your Google Gemini API key. (Note: You only need one Gmail OAuth2 credential. n8n will use this single connection for both fetching your inbox and sending the final summary email). Config node & Fallback: Open ⚙️ Set Configuration and fill in your target email, Slack user ID, and WhatsApp number. (Optional: Open the "Gmail: Send Error Alert" node at the bottom and replace PASTE_FALLBACK_EMAIL_HERE with a backup admin email, ensuring you get crash reports even if the config node fails). Toggle channels: Set each receiveOverviewIn... field to true for the platforms you want to activate. Test: Click Test Workflow — a sample email with a dummy password is pinned to the trigger so you can see the PII scrubber and summary in action immediately. Activate: Switch the workflow to Active and your first briefing runs on schedule. Requirements n8n version 1.0+ Google Gemini API key > Note on Model Selection: This template uses gemini-2.5-flash, which is currently in preview. If it is unavailable in your region or you encounter rate limits, simply open the Generate AI Summary node and switch the model to gemini-1.5-flash. API scopes: Gmail (readonly, send), Google Docs (write), Slack (chat:write), WhatsApp Business Customization ideas Change email scope:** Edit the labels in the Gmail fetch node to include Promotions, Updates, or a specific sender. Filter by project:** Enable filterBySubject or filterBySender in the config node to build a summary focused on one client or topic. Adjust the tone:** Edit the prompt in the Build LLM Prompt node to make summaries more detailed, more concise, or in a different language. Change the schedule:** Update the Schedule Trigger to run twice a day, weekly, or on any cadence you need. Use cases Start every morning with a 60-second read of everything that matters in your inbox. Keep a searchable daily email log in Google Docs without lifting a finger. Stay on top of client communications even when you're away from your inbox. Safely use AI for email management without exposing passwords or OTPs to external model providers.

Build a fully functional AI chatbot for any website using Retrieval-Augmented Generation (RAG). This workflow automatically crawls and indexes your entire site into a Qdrant vector database, then powers a conversational chatbot that searches your content to answer user questions — and escalates unresolved issues to your support team via Gmail. How it works Indexing Pipeline A Code node defines which root domains to crawl Firecrawl maps every link across those domains before scraping begins, giving you full visibility of what will be indexed without wasting credits Duplicate URLs are removed across all domains before any scraping starts Each unique page is scraped individually and returned as clean markdown Content is chunked into overlapping segments using a Recursive Character Text Splitter (1000 characters, 200 overlap) to preserve context at chunk boundaries Mistral's codestral-embed-2505 model converts each chunk into a vector embedding All embeddings are stored in Qdrant Cloud in batches of 100 A Wait node paces the loop to avoid hitting API rate limits on large sites AI Chatbot A public Chat Trigger receives messages and generates an embeddable URL for your website GPT-4o-mini processes each message with a 10-message memory window for natural conversation The AI Agent searches the Qdrant vector store only when a question requires it, retrieving the top 3 most relevant chunks per query When it cannot resolve an issue, it collects the user's email, writes a summary, confirms with the user, then sends it via Gmail How to use Add all required credentials in n8n Settings > Credentials Create a Qdrant Cloud collection (1536 dimensions, Cosine distance) Update the collection name in both Qdrant Vector Store nodes Open the "set urls to scrape" Code node and replace the placeholder URLs with your own site's root domains Update the Gmail tool with your support inbox address Run the indexing pipeline manually using the Run Indexing trigger Once indexing is complete, activate the workflow and test via Open Chat Embed the chat trigger URL on your website Requirements Firecrawl** — for site mapping and scraping (firecrawl.dev) Mistral Cloud** — for embeddings in both indexing and retrieval (console.mistral.ai) Qdrant Cloud** — for vector storage and semantic search (cloud.qdrant.io) OpenAI** — for the GPT-4o-mini chat model (platform.openai.com) Gmail OAuth2** — for support email escalation Customising this workflow Swap GPT-4o-mini for any chat model supported by n8n's LangChain nodes including Gemini, Claude, or Mistral Change the embedding model — if you do, delete and recreate the Qdrant collection with the correct dimensions and re-run indexing Add more URLs to the Code node array to index additional domains Adjust chunk size in the Text Splitter for denser or shorter content Increase the retrieval limit from 3 if answers feel incomplete Replace Gmail with Slack, Zendesk, or any other escalation tool Update the AI Agent system prompt to match your own website and brand voice

This template enriches a lead list by analyzing each contact’s LinkedIn activity and auto-generating a single personalized opening line for cold outreach. Drop a spreadsheet into a Google Drive folder → the workflow parses rows, fetches LinkedIn content (recent post or profile), uses an LLM to craft a one-liner, writes the result back to Google Sheets, and sends a Telegram summary. ⸻ Good to know • Works with two paths: • Recent post found → personalize from the latest LinkedIn post. • No recent post → personalize from profile fields (headline, about, current role). • Requires valid Apify credentials for LinkedIn scrapers and LLM keys (Anthropic and/or OpenAI). • Costs depend on the LLM(s) you choose and scraping usage. • Replace all placeholders like [put your token here] and [put your Telegram Bot Chat ID here] before running. • Respect the target platform’s terms of service when scraping LinkedIn data. What this workflow does Trigger (Google Drive) – Watches a specific folder for newly uploaded lead spreadsheets. Download & Parse – Downloads the file and converts it to structured items (first name, last name, company, LinkedIn URL, email, website). Batch Loop – Processes each row individually. Fetch Activity – Calls Apify LinkedIn Profile Posts (latest post) and records current date for recency checks. Recency Check (LLM) – An OpenAI node returns true/false for “post is from the current year.” Branching • If TRUE → AI Agent (Anthropic) crafts a single, natural reference line based on the recent post. • If FALSE → Apify LinkedIn Profile → AI Agent (Anthropic) crafts a one-liner from profile data (headline/about/current role). Write Back (Google Sheets) – Updates the original sheet by matching on email and writing the personalization field. Notify (Telegram) – Sends a brief completion summary with sheet name and link. Requirements • Google Drive & Google Sheets connections • Apify account + token for LinkedIn scrapers • LLM keys: Anthropic (Claude) and/or OpenAI (you can use one or both) • Telegram bot for notifications (bot token + chat ID) How to use Connect credentials for Google, Apify, OpenAI/Anthropic, and Telegram. Set your folder in the Google Drive Trigger to the one where you’ll drop lead sheets. Map sheet columns to the expected headers (e.g., First Name, Last Name, Company Name for Emails, Person Linkedin Url, Email, Website). Replace placeholders ([put your token here], [put your Telegram Bot Chat ID here]) in the respective nodes. Upload a test spreadsheet to the watched folder and run once to validate the flow. Review results in your sheet (new personalization column) and check Telegram for the completion message. Setup Connect credentials - Google Drive/Sheets, Apify, OpenAI and/or Anthropic, Telegram. Configure the Drive trigger - Select the folder where you’ll upload your lead sheets. Map columns - Ensure your sheet has: First Name, Last Name, Company Name for Emails, Person Linkedin Url, Email, Website. Replace placeholders - In HTTP nodes: Bearer [put your token here]. In Telegram node: [put your Telegram Bot Chat ID here] (Optional) Adjust the recency rule - Current logic checks for current-year posts; change the prompt if you prefer 30-day windows. How to use Upload a test spreadsheet to the watched Drive folder. Execute the workflow once to validate. Open your Google Sheet to see the new personalization column populated. Check Telegram for the completion summary. Customizing this template • Data sources: Add company news, website content, or X/Twitter as fallback signals. • LLM choices: Use only Anthropic or only OpenAI; tweak temperature for tone. • Destinations: Write to a CRM (HubSpot/Salesforce/Airtable) instead of Sheets. • Notifications: Swap Telegram for Slack/Email/Discord. Who it’s for • Sales & SDR teams needing authentic, scalable personalization for cold outreach. • Lead gen agencies enriching spreadsheets with ready-to-use openers. • Marketing & growth teams improving reply rates by referencing real prospect activity. Limitations & compliance • LinkedIn scraping may be rate-limited or blocked; follow platform ToS and local laws. • Costs vary with scraping volume and LLM usage. Need help customizing? Contact me for consulting and support: LinkedIn