Workflow Templates

Sort invoice PDFs from Gmail to Google Drive and Google Sheets Who’s it for Freelancers, finance teams, and small businesses that receive invoice PDFs by email and want them automatically saved to Google Drive and logged in Google Sheets—without manual downloading or copy-pasting. How it works / What it does This workflow watches your Gmail inbox for unread messages that match an invoice pattern (e.g., subject:invoice filename:pdf). For each email, it checks for attachments, uploads each PDF to a chosen Google Drive folder, and appends a new row to a Google Sheet with useful metadata: received time, sender, subject, filename, Drive link, and IDs. Finally, it marks the original email as read to avoid duplicates. How to set up Open the Config (Set) node and fill in: drive_folder_id (or leave blank for root) spreadsheet_id (from the Sheet URL) sheet_name (e.g., Invoices) Connect credentials for Gmail, Google Drive, and Google Sheets in each node. Adjust the Gmail search query if needed (language/vendor terms). Run once manually to verify data mapping, then activate. Requirements n8n with valid credentials for Gmail, Google Drive, and Google Sheets. A Google Sheet with appropriate headers (or let the workflow write new columns). How to customize the workflow Replace Gmail with IMAP or Microsoft Outlook if you don’t use Gmail; remove the “mark as read” step accordingly. Add parsing (e.g., extract invoice totals or vendor names via PDF/AI nodes) before the Sheets step. Route based on vendor: create subfolders dynamically in Drive and write to different tabs. Notify your team by adding Slack/Email nodes after logging to Sheets.

🧠 Overview This multi-agent n8n automation simulates a high-functioning marketing team. A strategic CMO agent receives your chat-based input, decides which specialist is best for the task, and delegates accordingly. Each specialist (copywriter, SEO expert, brand strategist, etc.) operates independently using fast, cost-effective GPT-4.1-mini models—resulting in parallel task execution and full-funnel marketing output with minimal human input. ⚙️ How It Works A chat message trigger listens for input (e.g. “Write a full email funnel for our SaaS launch”). The CMO Agent (powered by OpenAI O3) reads the message and determines intent, strategy, and needed outputs. It dynamically delegates tasks to the correct AI agent: Copywriter Agent Facebook Ads Specialist SEO Content Writer Email Marketer Social Media Manager Brand Voice Specialist Each agent uses a dedicated GPT-4.1-mini model to produce results instantly. Final content is returned to the user or passed along for integration with your CMS, ad platforms, or CRM. 🧰 Tools Used n8n** – Orchestrates the entire agent communication and routing logic OpenAI O3** – Advanced strategic reasoning (CMO Agent) OpenAI GPT-4.1-mini** – Fast and cost-efficient for specialist agents LangChain Nodes** – For multi-agent thinking and tool-based execution 🚀 Quick Start Import Workflow: Load the provided .json into your n8n instance Set Credentials: Add your OpenAI API key under “OpenAI Account” Deploy Webhook: Use the “When Chat Message Received” trigger Test It: Ask a question like: > “Generate a 7-day onboarding email sequence for a weight loss app” Watch the Agents Collaborate! 👩‍💼 Meet Your AI Marketing Team | Agent | Purpose | Model | Output | |-------|---------|-------|--------| | 🧠 CMO Agent | Strategy, delegation, and task routing | O3 | Central brain | | ✍️ Copywriter Agent | Website copy, CTAs, product descriptions | GPT-4.1-mini | Fast, human-like copy | | 📱 Facebook Ads Copywriter | Ad headlines, angles, A/B tests | GPT-4.1-mini | Platform-specific ad copy | | 🔍 SEO Writer | Blog posts, keyword-rich content | GPT-4.1-mini | Long-form content | | 📧 Email Specialist | Sequences, newsletters, welcome flows | GPT-4.1-mini | Funnel-ready emails | | 📲 Social Media Manager | Content calendars, posts, hashtags | GPT-4.1-mini | Cross-platform content | | 🎨 Brand Voice Specialist | Tone consistency, style guides | GPT-4.1-mini | On-brand text | 💡 Use Cases Product Launches:** Strategy → Landing Page → Emails → Social Posts Lead Nurture Funnels:** Segmented email campaigns with consistent tone Content Sprints:** Generate 30+ blog posts and socials in a day Ad Variations:** Create 20 ad angles in 30 seconds Brand Guidelines:** Enforce consistent messaging across departments 💸 Cost Optimization Use O3 sparingly—only for strategic tasks All specialist agents use GPT-4.1-mini for low-latency, high-efficiency generation Run agents in parallel to reduce wait times Add caching for repeat requests 🔧 Customization Tips Edit the tool prompts to match your brand’s style and niche Connect outputs to Google Sheets, Notion, Slack, or email tools Integrate with Zapier, Make.com, or your CRM for full automation 🔗 Connect With Me Website:** nofluff.online YouTube:** @YaronBeen LinkedIn:** Yaron Been 🏷️ Tags #n8n #OpenAI #MarketingAI #CMOagent #Automation #GPT4 #LangChain #NoCode #MarketingTeam #AIWorkflow #EmailMarketing #SEO #Copywriting #SocialMedia #DigitalMarketing #BrandVoice #AItools #MultiAgentSystem #ContentCreation #MarketingStrategy #ContentOps

Use cases are many: Automate Gmail tasks such as sending, replying, labeling, deleting, and fetching emails — all with AI assistance. Perfect for YouTubers managing viewer emails, sales teams handling inquiries, freelancers responding to client requests, or professionals keeping their inbox organized. Good to know At time of writing, each Gemini request is billed per token. See Gemini Pricing for updated details. The workflow uses Gmail labels (e.g., youtube-viewers, sales-inquiry, meeting-request, potential-clients, collaboration-requests) for classification — make sure these exist in your Gmail account. How it works Chat Trigger**: You interact with the agent via a chat interface (webhook). AI Agent**: Gemini-powered assistant interprets your instructions (send, reply, label, delete, fetch emails). Email Actions**: Based on your request, the assistant uses Gmail tools to act on emails (Send, Reply, Label, Delete, Get Many). Contact Lookup**: If only a name is provided, the agent checks Google Sheets for the matching email address. If not found, it prompts you to add it. Memory**: A buffer memory stores chat context so the assistant can maintain continuity across multiple interactions. Labeling**: Emails can be auto-labeled for better organization (e.g., client inquiries, meeting requests). How to use Send commands like: “Reply to John’s email with a follow-up about the project.” “Label Sarah’s email as potential-client.” “Delete the latest spam email.” The Gmail Agent will handle the request instantly and keep everything logged properly. Requirements Gmail account connected with OAuth2 credentials Google Gemini API key for AI processing Google Sheets for contact management Pre-created Gmail labels for organization Customising this workflow Add new Gmail labels for your workflow (e.g., Invoices, Support Tickets). Connect to a CRM (e.g., HubSpot, Notion, or Airtable) for syncing email data. Enhance AI replies with dynamic templates stored in Google Sheets. Extend chat commands to include batch actions (e.g., “Archive all emails older than 30 days”).

This workflow provides a way to capture detailed AI prompts using a voice note transcription service and then passes them on for completion to an AI agent. To preserve outputs in a knowledge management system, the AI response and the prompt are combined into one document that is created in a Nuclino collection (note: the Nuclino step is configured manually with a HTTP request node). How it works A webhook receives voice note data from Voicenotes.com containing the title and transcript The transcript is extracted and sent to an AI Agent powered by OpenRouter's Claude Sonnet model The AI generates a structured response in markdown format with Summary, Prompt, and Response sections The original prompt and AI response are merged and prepared for multiple outputs A Nuclino document is created via HTTP Request with the structured content A Slack notification is sent with the prompt, response, and Nuclino note URL Both the original prompt and AI response are archived in NocoDB for future reference How to use The webhook trigger can be configured to receive data from Voicenotes.com or any service that provides title and transcript data Replace the manual trigger with webhook, form, or other triggers as needed Customize the AI system message to change response format and behavior Configure Nuclino workspace and collection IDs for proper document organization Requirements OpenRouter account** for AI model access (Claude Sonnet) Nuclino account** and API token for document creation Slack workspace** with bot permissions for notifications NocoDB instance** for archiving (optional) Voicenotes.com account** for voice input (or alternative webhook source) Customising this workflow AI Models**: Switch between different OpenRouter models by changing the model parameter Response Format**: Modify the AI Agent system message to change output structure Documentation Platforms**: Replace Nuclino HTTP Request with other documentation APIs Notification Channels**: Add multiple Slack channels or other notification services Archive Storage**: Replace NocoDB with other database solutions Input Sources**: Adapt webhook to accept data from different voice note or transcription services Nuclino API The Nuclino API is documented here.

How it works time trigger using the cron format, every weekday at 5pm gets CentralStationCRM people updates of today checks for tag "Outreach" if true, sends message on gmail (predefine in node) waits 7 days, checks for answers alerts user if an answer is there if not, repeats process with second mail How to set up get credentials for CentralStationCRM, Slack and GMail set up the respective nodes with the credentials define text for your automated mails test without wait nodes

Publish a Reel on a Facebook Page 🧠 Good to Know > Set up n8n self-hosted instance using https://tino.vn/vps-n8n?affid=388 > Use the code ==VPSN8N== for up to 39% off. This workflow automatically uploads and publishes Facebook Reels using data from Google Sheets and video files stored in Google Drive. It runs on a schedule (every 30 minutes by default) and can be fully customized for your posting routine. ⚙️ How It Works Google Sheets provides the content data — video file ID, caption, and optional links. Google Drive hosts the actual video file (.mp4 format). The workflow initializes an upload session with the Facebook Graph API, uploads the video, and publishes it as a Reel on your Page. Finally, it updates your Google Sheet and adds a comment under the published Reel with your affiliate or product link. How to Use Open the template Google Sheet or make a copy: 👉 Template Sheet Fill out each row with: File ID → the ID of your video file from Google Drive File name → optional Caption → your post caption Link Share → optional Link post → leave empty (it will be filled after posting) Ensure your video file: .mp4 format shared folder on Google Drive that’s accessible to your connected account Add your Facebook Page ID and Page Access Token to the “info” node. (Learn how to get these here: Facebook Reels Workflow Guide) 📋 Requirements n8n instance (Self-hosted recommended):** Set up a self-hosted instance using https://tino.vn/vps-n8n?affid=388 Use the code VPSN8N for up to 39% off. Facebook Page** with publishing permissions Page Access Token** (with pages_manage_posts, pages_read_engagement) Google Drive* and *Google Sheets** account connected to n8n Video files in .mp4 format, under the 1GB upload limit 🎨 Customizing This Workflow Change schedule:* Adjust the *Schedule Trigger** node (e.g., every 2 hours or only during business hours). Track post links:** Add a node to fetch the permalink_url from the Graph API and update it in your sheet. Auto-comment control:** Modify or remove the “Create comment post” node to suit your campaign style. Improve security:* Replace hardcoded tokens with *n8n credentials, **secrets, or a Data Table lookup. This structure keeps your automation scalable, secure, and easy to adapt for multi-page or multi-brand use.

Task-based Assignee billing via Time Tracking This workflow automates billing by scanning a target Dartboard on schedule, aggregating time logs from completed tasks, cross‑referencing assignee rates in Google Sheets, calculating total pay, and updating the sheet with final billable hours and amounts. Who's it for Individuals, agencies, companies, and project managers automating payroll or client invoicing from task data. How to setup Link your Dart and Google accounts. Replace the dummy ID in the List tasks node with your actual target Dartboard ID. Set your preferred run frequency (e.g., Weekly). Create a Google Sheet with these exact headers: Name, HourlyRate, TotalHours, TotalPay, DateCalculated. Connect the Sheet nodes to your file. Pre-fill Name (matching Dart Assignees exactly) and HourlyRate in your Google Spreadsheet. Optional: Add a last header column in the sheet as a Status header to track if the bill is paid or pending. Customizing the workflow Choose your AI model for your AI time tracking and assignee scanner Use your own google sheet account and target spreadsheet document

🧠 AI Telegram Customer Support Assistant + Lead Manager This n8n workflow provides a fully automated AI-driven customer support assistant connected to Telegram, with built-in lead management, knowledge-base querying, and context-aware answers. ⭐ What this workflow does Receives user messages from Telegram Logs all incoming/outgoing messages into a Data Table Checks if a lead exists for the user (via chat_id) Creates new leads automatically if needed Builds an AI-ready context (user info + lead info + latest message) Uses the AI Agent to answer questions using: FAQ database Services table (programs, prices, descriptions) Settings table (agency info) Lead update tool Sends a natural, friendly reply back to Telegram Updates leads in real time (name, phone, email, notes…) 📦 Required Data Tables (to be created by the user) leads Stores all user information (full_name, phone, email, etc.) services List of available programs/services with prices, duration, category. faq Frequently asked questions with answers and optional tags. settings Company/agency details used by the assistant. chat_logs Logs all messages exchanged with users (user + bot). 🔧 Required Credentials Telegram Bot API Key AI Model Credential (OpenAI, OpenRouter, Groq…) No other credentials required. 🚀 How to use it Import the workflow into your n8n instance Create the required Data Tables as defined inside the Sticky Notes Add your credentials (Telegram + AI Model) Customize the prompt to match your business Activate the workflow — you're ready to go! 💡 Suitable for: Travel agencies Customer support chatbots Lead qualification + automation AI knowledge-based assistants Telegram-first businesses

🛠️ Landing-Page Roast & CRO Ideas Bot – Quick Guide What this workflow does Takes any public landing-page URL. Scrapes the page content. Uses Gemini 2.5-pro to • Roast the page (friendly but brutally honest) • Give 10 high-impact, 2024-ready CRO ideas – all in Turkish, max 3 000 characters. Sends the result back to you on Telegram. Two ways to trigger it Web form • Open the form titled “Conversion Rate Optimizer.” • Paste your landing-page URL(with https or http in front of it). • Click Submit. Telegram (fastest) • Send the URL in a DM to @MertSiteRaporBot. • Forgot the “https://”? No worries—the bot adds it automatically. Behind the scenes • Code node normalises the URL. • HTTP Request scrapes the page HTML. • AI Agent (Gemini) produces the Roast + Recommendations. • Telegram node sends the formatted reply to you. Usage tips • One URL per request. • Page must be publicly accessible (no login). • Very long pages may be trimmed to fit model limits. • Output language is always Turkish.

Automatically Import Research Papers using DOI URL from Telegram to Zotero This workflow allows you to automatically import research papers into your Zotero library by simply sending a DOI link through Telegram. It fetches metadata from reliable sources such as Crossref, DataCite, and Unpaywall, enriches the bibliographic details, attaches the best available full-text PDF when possible, and generates a concise summary of the abstract using an LLM before sending it back to Telegram. ✨ Key Features Telegram Integration** – Send a DOI link via Telegram bot to trigger the workflow. DOI Parsing** – Automatically detects and extracts DOI or arXiv identifiers from user messages. Metadata Fetching* – Retrieves bibliographic information from *Crossref, **DataCite, and Unpaywall. Smart PDF Finder** – Locates the best available PDF (publisher link, open access, or arXiv). Zotero Integration** – Creates a Zotero parent item with complete metadata and attaches the PDF link when available. Abstract Summarization** – Uses a basic LLM chain to generate a short and clear summary of the abstract. Telegram Feedback** – Sends the title, URL, and abstract summary back to the user in Telegram. 🔑 Required Credentials Telegram API** – To connect the workflow with your Telegram bot. Zotero API Key** – To create and update items in your Zotero library. OpenRouter API Key** – To enable the LLM for generating abstract summaries. (Optional) Email for Unpaywall API requests (recommended for stable access). 💡 Benefits Save time by automating manual research paper imports. Ensure high-quality metadata by combining multiple sources (Crossref, DataCite, Unpaywall). Get instant summaries of abstracts directly in Telegram for quick understanding. Keep your Zotero library organized with accurate titles, abstracts, authors, and tags. Quickly attach available full-text PDFs without manual searching. Improve your academic workflow by managing references and summaries directly from Telegram.

The problem Ever attend a networking event and find yourself taking screenshots of people's LinkedIn? Sounds counter-intuitive because you are connecting on LinkedIn. But you find it hard to keep track of everyone you've met. You also don't want to miss diligently updating your CRM with details and insights. *The solution * There's no need for yet another app. Continue taking screenshots. Just share them on a 2-field only Google Form: screenshot + your quick notes about the person. Create a shortcut to the Google Form link on your phone homescreen. Voila! You have app-like access without the need for an app. Once you submit with just these 2 pieces of info, AI parses the image AND crafts a follow-up message. Within minutes! Just open your spreadsheet to have all that information consolidated - automatically - for your review. Promote yourself from do-er to manager. Who should use it? Anyone really. If you find yourself meeting people but want to be more meticulous or efficient staying on top, use this. How to set it up Time: ~10 minutes end-to-end. Import the provided workflow JSON in n8n. Connect credentials: Google Drive (read), Google Sheets (write), OpenAI. Configure key information: Google Sheets and relevant columns Configure Open AI models based on your cost/ efficiency requirements Confirm column headers in your Sheet match the variables (or update the variables). Test with one screenshot. Pro-tip: Add that Google Form link as a shortcut on your phone's home screen. Get app-like convenience without downloading yet another app.

Try It Out! This workflow builds a Telegram-based Solana (SOL/USDT) Multi-Timeframe AI Market Analyzer that automatically pulls live candlestick data for SOL/USDT, runs structured multi-timeframe technical analysis (1-minute, 5-minute, 1-hour) through an AI Agent, and posts a professional, JSON-structured analysis + trading recommendation straight to your Telegram chat. It combines on-chain / market data aggregation, LLM-driven interpretation, and instant Telegram reporting — giving you concise, actionable market intelligence every hour. How It Works Hourly Trigger – The workflow runs once per hour to pull fresh market data. Market Data Fetch – Three HTTP requests gather candlesticks from CryptoCompare: 1-minute (last 60 candles) 5-minute aggregate (last 60 aggregated candles) 1-hour (last 60 candles) Merge & Transcribe – The three feeds are merged and a lightweight code node extracts: symbol, current price, arrays for data_1m, data_5m, data_1h. AI Agent Analysis – The LLM (configured via your model node) receives the merged payload and runs a structured multi-timeframe technical analysis, returning a strict JSON report containing: Per-timeframe analysis (momentum, volume, S/R,MA, volatility) Market structure / confluence findings Trading recommendation (action, entry, stop, TPs, position sizing) A final disclaimer Parse AI Output – Extracts the JSON block from the agent’s reply and validates/parses it for downstream formatting. Telegram Reporting – Sends two nicely formatted Telegram messages: Multi-timeframe breakdown (1m / 5m / 1h) Market structure + Trading Recommendation (TPs, SL, position size, disclaimer) How to Use Import the workflow into your n8n workspace (or replicate the nodes shown in the JSON). Add credentials: CryptoCompare API Key — for reliable candlestick data. LLM model credentials — e.g., Google Gemini / OpenAI, configured in the LangChain/LM node. Telegram Bot Token & Chat ID — to send messages. (Optional) AFK Crypto API key if you want to enrich data with wallet info later. Node mapping & endpoints: Fetch_1m → GET https://min-api.cryptocompare.com/data/v2/histominute?fsym=SOL&tsym=USDT&limit=60 Fetch_5m → GET https://min-api.cryptocompare.com/data/v2/histominute?fsym=SOL&tsym=USDT&limit=60&aggregate=5 Fetch_1h → GET https://min-api.cryptocompare.com/data/v2/histohour?fsym=SOL&tsym=USDT&limit=60 Merge → combine the three responses into a single payload. Transcribe (code) → extract last close as current price and attach the arrays. AI Agent → pass the structured prompt (system message instructs exact JSON structure). Parse AI Output → extract the json ... block and JSON.parse it. Telegram nodes → format and send two messages (timeframes and recommendation). Adjust analysis frequency: default is hourly — change the Schedule Trigger node as desired. Deploy and activate: the workflow will post an AI-driven SOL/USDT market analysis to your Telegram hourly. (Optional) Extend This Workflow Add price / orderbook enrichment (e.g., AFK price endpoints or exchange orderbook) to improve context. Add wallet exposure checks (AFK wallet balances) to tailor position sizing suggestions. Store AI reports in Notion / Google Sheets for historical auditing and backtesting. Add alert filtering to only post when the LLM flags high-confidence signals or confluence across timeframes. Expose Telegram commands to request on-demand analysis (e.g., /analyze now 5m). Add risk management logic to convert LLM recommendation into automated orders (careful — requires manual review and stronger safety controls). Safety Mechanisms Explicit system prompt — forces AI to output only the exact JSON structure to avoid free-form text parsing errors. JSON parser node — validates the agent response and throws if malformed before any downstream action. Read-only market analysis — the workflow only reports by default (no auto-trading), reducing operational risk. Credentials gated — ensure LLM and Telegram credentials are stored securely in n8n. Disclaimer — every report includes a legal/financial disclaimer from the agent. Requirements CryptoCompare API Key (for minute/hour candlesticks) LLM model credentials (Google Gemini / OpenAI / other supported model in your LangChain node) Telegram Bot Token + Chat ID (where analysis messages are posted) Optional: AFK Crypto API key if you plan to add wallet/position context n8n instance with: HTTP Request, Code, Merge, LangChain/Agent, and Telegram nodes enabled AFK / External APIs Used CryptoCompare Candles: GET https://min-api.cryptocompare.com/data/v2/histominute?fsym=SOL&tsym=USDT&limit=60 (1m) GET https://min-api.cryptocompare.com/data/v2/histominute?fsym=SOL&tsym=USDT&limit=60&aggregate=5 (5m) GET https://min-api.cryptocompare.com/data/v2/histohour?fsym=SOL&tsym=USDT&limit=60 (1h) Telegram Bot API – via n8n Telegram node. LLM / LangChain – your chosen LLM provider (configured in the workflow). Summary The Solana (SOL/USDT) Multi-Timeframe AI Market Analyzer (Telegram) gives you hourly, professional multi-timeframe technical analysis generated by an LLM agent using real candlestick data from CryptoCompare. It combines the speed of automated data collection with the structure and reasoning of an AI analyst, delivering clear trading recommendations and a timestamped analysis to your Telegram chat — ideal for traders who want reliable, concise market intelligence without manual charting. Our Website: https://afkcrypto.com/ Check our blogs: https://www.afkcrypto.com/blog