by dongou
Fetch user-specific research papers from arXiv on a daily schedule, process and structure the data, and create or update entries in a Notion database, with support for data delivery Paper Topic**: single query keyword Update Frequency**: Daily updates, with fewer than 20 entries expected per day Tools**: Platform: n8n, for end-to-end workflow configuration AI Model: Gemini-2.5-Flash, for daily paper summarization and data processing Database: Notion, with two tables — Daily Paper Summary and Paper Details Message: Feishu (IM bot notifications), Gmail (email notifications) 1. Data Retrieval arXiv API The arXiv provides a public API that allows users to query research papers by topic or by predefined categories. arXiv API User Manual Key Notes: Response Format: The API returns data as a typical Atom Response. Timezone & Update Frequency: The arXiv submission process operates on a 24-hour cycle. Newly submitted articles become available in the API only at midnight after they have been processed. Feeds are updated daily at midnight Eastern Standard Time (EST). Therefore, a single request per day is sufficient. Request Limits: The maximum number of results per call (max_results) is 30,000, Results must be retrieved in slices of at most 2,000 at a time, using the max_results and start query parameters. Time Format: The expected format is [YYYYMMDDTTTT+TO+YYYYMMDDTTTT], TTTT is provided in 24-hour time to the minute, in GMT. Scheduled Task Execution Frequency**: Daily Execution Time**: 6:00 AM Time Parameter Handling (JS)**: According to arXiv’s update rules, the scheduled task should query the previous day’s (T-1) submittedDate data. 2. Data Extraction Data Cleaning Rules (Convert to Standard JSON) Remove Header Keep only the 【entry】【/entry】 blocks representing paper items. Single Item Each 【entry】【/entry】 represents a single item. Field Processing Rules 【id】【/id】 ➡️ id Extract content. Example: 【id】http://arxiv.org/abs/2409.06062v1【/id】 → http://arxiv.org/abs/2409.06062v1 【updated】【/updated】 ➡️ updated Convert timestamp to yyyy-mm-dd hh:mm:ss 【published】【/published】 ➡️ published Convert timestamp to yyyy-mm-dd hh:mm:ss 【title】【/title】 ➡️ title Extract text content 【summary】【/summary】 ➡️ summary Keep text, remove line breaks 【author】【/author】 ➡️ author Combine all authors into an array Example: [ "Ernest Pusateri", "Anmol Walia" ] (for Notion multi-select field) 【arxiv:comment】【/arxiv:comment】 ➡️ Ignore / discard 【link type="text/html"】 ➡️ html_url Extract URL 【link type="application/pdf"】 ➡️ pdf_url Extract URL 【arxiv:primary_category term="cs.CL"】 ➡️ primary_category Extract term value 【category】 ➡️ category Merge all 【category】 values into an array Example: [ "eess.AS", "cs.SD" ] (for Notion multi-select field) Add Empty Fields github huggingface 3. Data Processing Analyze and summarize paper data using AI, then standardize output as JSON. Single Paper Basic Information Analysis and Enhancement Daily Paper Summary and Multilingual Translation 4. Data Storage: Notion Database Create a corresponding database in Notion with the same predefined field names. In Notion, create an integration under Integrations and grant access to the database. Obtain the corresponding Secret Key. Use the Notion "Create a database page" node to configure the field mapping and store the data. Notes "Create a database page"** only adds new entries; data will not be updated. The updated and published timestamps of arXiv papers are in UTC. Notion single-select and multi-select fields only accept arrays. They do not automatically parse comma-separated strings. You need to format them as proper arrays. Notion does not accept null values, which causes a 400 error. 5. Data Delivery Set up two channels for message delivery: EMAIL and IM, and define the message format and content. Email: Gmail GMAIL OAuth 2.0 – Official Documentation Configure your OAuth consent screen Steps: Enable Gmail API Create OAuth consent screen Create OAuth client credentials Audience: Add Test users under Testing status Message format: HTML (Model: OpenAI GPT — used to design an HTML email template) IM: Feishu (LARK) Bots in groups Use bots in groups
by Athanasios
AI Interior Design Assistant: Your Digital Design Partner What This System Does This n8n workflow transforms your Telegram into a professional interior design studio powered by artificial intelligence. Send a photo of furniture or a room space, and watch as the system intelligently catalogs items, documents spaces, and generates stunning custom interior designs tailored to your vision. The Magic Behind the Scenes Smart Image Recognition When you upload a photo, the system immediately springs into action with sophisticated image analysis: Furniture Detection**: Spots individual pieces like sofas, chairs, tables, and lamps with catalog-precision accuracy Room Analysis**: Identifies complete spaces, architectural features, lighting conditions, and existing design elements Style Classification**: Determines design styles from modern minimalist to traditional classic Material Recognition**: Identifies wood types, fabric textures, metal finishes, and color palettes Intelligent Database Management The workflow maintains three interconnected databases that work like a professional design firm's catalog system: Furniture Catalog (catalog_products) Comprehensive product details including style, materials, dimensions, and compatibility Professional descriptions written for interior designers Searchable tags for quick design matching High-quality image storage for visual reference Room Documentation (rooms) Detailed space analysis including size, style, and architectural features Color palette documentation and lighting assessment Existing furniture inventory for design planning Room-specific design recommendations Design Portfolio (ai_generated_images) Archive of all AI-generated interior designs Original prompts and design descriptions Searchable by style, room type, or specific elements Ready for client presentations or further modifications AI-Powered Design Generation The system's crown jewel is its ability to create stunning interior visualizations: Contextual Understanding: Combines room characteristics with catalog products to create realistic design scenarios Professional Prompting: Generates detailed, interior-design-specific prompts that result in high-quality, commercially viable designs Style Consistency: Maintains design coherence across different elements while respecting user preferences Modification Capabilities: Can reference and modify previous designs, allowing for iterative improvements The User Experience Journey Scenario 1: Building Your Furniture Catalog Upload: Send photos of furniture pieces via Telegram Analysis: AI examines each piece, identifying style, materials, dimensions, and design era Cataloging: Items are professionally documented with searchable metadata Confirmation: Receive detailed catalog entries for each piece Scenario 2: Documenting Your Spaces Room Photos: Share images of your living spaces Space Analysis: AI assesses room size, style, lighting, and architectural features Documentation: Complete room profiles are created for design planning Inventory: Existing furniture and design elements are noted Scenario 3: Creating Custom Designs Design Request: Ask for specific interior modifications or new layouts Smart Matching: System pulls relevant items from your catalog and room data AI Generation: Gemini 2.5 Flash creates photorealistic interior designs Instant Delivery: Receive professional-quality visualizations via Telegram Scenario 4: Design Evolution Reference Previous Work: Mention earlier designs you want to modify Contextual Modification: AI understands your reference and applies new changes Enhanced Generation: Creates updated designs building on previous concepts Continuous Improvement: Iterate until the design matches your vision Technical Sophistication Multi-AI Coordination OpenAI GPT-4**: Handles complex reasoning, database operations, and user interaction Google Gemini 2.5 Flash**: Specializes in high-quality image generation with interior design expertise Intelligent Routing**: Automatically determines whether to catalog, document, or generate based on context Professional Data Structure The database schema reflects real interior design workflows: Industry-standard categorization systems Professional terminology and measurements Design compatibility matrices Style and era classifications used by actual designers Seamless Integration Telegram Interface**: No app downloads or complex interfaces - just send photos and text Cloud Storage**: All images stored in Supabase with public URLs for easy access Real-time Processing**: Immediate feedback and rapid design generation Persistent Memory**: Everything is saved and searchable for future reference Why This Matters This workflow bridges the gap between professional interior design tools and accessible consumer technology. It provides: For Design Professionals: A powerful cataloging and visualization tool that streamlines client presentations and design iteration For Homeowners: Professional-level design capability without the cost or complexity of traditional design software For Businesses: A scalable solution for furniture visualization, space planning, and customer engagement The Innovation Factor Unlike simple design apps that work with generic templates, this system: Learns your specific furniture and spaces Maintains design continuity across projects Provides professional-quality outputs Scales from single rooms to complete home designs Integrates seamlessly into your daily communication workflow The result is a design assistant that feels less like software and more like having a professional interior designer available 24/7 through your phone. Future Possibilities This foundation supports expansion into: Room dimension calculations and space optimization Integration with furniture retailers for purchase links 3D room modeling and virtual reality previews Style preference learning and automated suggestions Multi-user collaboration for design teams The workflow represents a new paradigm where AI doesn't replace human creativity but amplifies it, making professional design capabilities accessible to anyone with a smartphone and an imagination.
by Mychel Garzon
AI-Powered CV Feedback & Fit Score This workflow uses AI to automatically analyze a candidate’s CV against any job posting. It extracts key skills, requirements, and gaps, then generates a clear fit summary, recommendations, and optimization tips. Candidates also receive a structured email report, helping them improve their CV and focus on the right roles. No more guesswork, the workflow delivers objective, AI-powered career insights in minutes. Benefits • Automated CV analysis: Instantly compare your CV with any job description. • Clear recommendations: Get a fit score (1–10) plus “Apply,” “Consider,” or “Not a fit.” • Actionable feedback: See missing skills and concrete optimization tips. • Email reports: Candidates receive a professional summary directly in their inbox. Target Audience • Job seekers • Career coaches and recruiters • HR teams evaluating candidate job alignment • Tech bootcamps and training programs Required APIs • Google Gemini API (AI analysis) • Email credentials (send candidate reports) Easy Customization • Fit score logic: Adjust thresholds for “Apply,” “Consider,” and “Not a fit.” • Email templates: Personalize branding, tone, or add follow-up resources. • Delivery channels: Add Slack, Teams, or WhatsApp nodes for real-time feedback. • Language detection: Extend to more languages by adding translation nodes.
by Rahul Joshi
Description: Keep your API documentation accurate and reliable with this n8n automation template. The workflow automatically tests your FAQ content related to authentication and rate limits, evaluating each answer using Azure OpenAI GPT-4o-mini for completeness, edge-case coverage, and technical clarity. It logs all results to Google Sheets, scores FAQs from 0–10, and sends Slack alerts when low-quality answers are detected. Ideal for API teams, developer relations managers, and technical writers who want to maintain high-quality documentation with zero manual review effort. ✅ What This Template Does (Step-by-Step) ▶️ Manual Trigger or On-Demand Run Start the evaluation anytime you update your FAQs — perfect for regression testing before documentation releases. 📖 Fetch FAQ Q&A from Google Sheets Reads FAQ questions and answers from your designated test sheet (columns A:B). Each Q&A pair becomes a test case for AI evaluation. 🤖 AI Evaluation via GPT-4o-mini Uses Azure OpenAI GPT-4o-mini to evaluate how well each FAQ covers critical aspects of API authentication and rate limiting. The AI provides a numeric score (0–10) and a short explanation. 🔍 Parse & Format AI Results Extracts structured JSON data (Question, Score, Explanation, Timestamp) and prepares it for reporting and filtering. 💾 Save Evaluation to Google Sheets Appends all results to a Results Sheet (A:D) — creating a running history of FAQ quality audits. ⚠️ Filter for Low-Scoring FAQs Identifies any FAQ with a score below 7, flagging them as needing review or rewrite. 🔔 Send Slack Alerts for Weak Entries Posts an alert message in your chosen Slack channel, including: The question text Score received AI’s explanation Link to the full results sheet This ensures your documentation team can quickly address weak or incomplete FAQ answers. 🧠 Key Features 🤖 AI-powered FAQ quality scoring (0–10) 📊 Automated tracking of doc health over time 📥 Seamless Google Sheets integration for results storage ⚙️ Slack notifications for underperforming FAQs 🧩 Ideal for continuous documentation improvement 💼 Use Cases 📘 Validate FAQ accuracy before API documentation updates ⚡ Auto-test new FAQ sets during content refresh cycles 🧠 Ensure API rate limit and auth topics cover all edge cases 📢 Alert documentation owners about weak answers instantly 📦 Required Integrations Google Sheets API – for reading and storing FAQs and test results Azure OpenAI (GPT-4o-mini) – for evaluating FAQ coverage and clarity Slack API – for sending quality alerts and notifications 🎯 Why Use This Template? ✅ Ensures API FAQ accuracy and completeness automatically ✅ Replaces tedious manual content reviews with AI scoring ✅ Builds an ongoing record of documentation improvements ✅ Keeps technical FAQs consistent, relevant, and developer-friendly
by Billy Christi
Who is this for? This workflow is perfect for: Support teams and customer service departments managing Jira tickets Team leads and managers who need daily visibility into ticket resolution progress Organizations wanting to automate ticket reporting and communication IT departments seeking to streamline support ticket summarization and tracking What problem is this workflow solving? Manual ticket review and reporting is time-consuming and often lacks comprehensive analysis. This workflow solves those issues by: Automating daily ticket analysis** by fetching, analyzing, and summarizing all tickets created each day Providing intelligent summaries** using AI to extract key insights from ticket descriptions, comments, and resolutions Streamlining communication** by automatically sending formatted daily reports to stakeholders Saving time** by eliminating manual ticket review and report generation What this workflow does This workflow automatically fetches daily Jira tickets, analyzes them with AI, and sends comprehensive summaries via email to keep your team informed about support activities. Step by step: Schedule Trigger runs the workflow automatically at your chosen interval (or manual trigger for testing) Set Project Key defines the Jira project to monitor (default: SUP project) Get All Tickets from the specified project created today Split Out extracts individual ticket data including key, summary, and description Loop Tickets processes each ticket individually through batch processing Get Comments from Ticket retrieves all comments and conversations for complete context Merge combines ticket data with associated comments for comprehensive analysis Ticket Summarizer (AI Agent) uses OpenAI GPT-5 to generate professional summaries and proposed solutions Set Output structures the AI analysis into standardized JSON format Aggregate collects all processed ticket summaries into a single dataset Format Body creates a readable email format with direct Jira ticket links Send Ticket Summaries delivers the daily report via Gmail How to set up Connect your Jira account by adding your Jira Software Cloud API credentials to the Jira nodes Add your OpenAI API key to the OpenAI Chat Model node for AI-powered ticket analysis Configure Gmail credentials for the Send Ticket Summaries node to deliver reports Update the recipient email in the "Send Ticket Summaries" node to your desired recipient Adjust the project key in the "Set Project Key" node to match your Jira project identifier Configure the schedule trigger to run daily at your preferred time for automatic reporting Customize the JQL query in Jira nodes to filter tickets based on your specific requirements Test the workflow using the manual trigger to ensure proper ticket fetching and AI analysis Review email formatting in the "Format Body" node and adjust as needed for your reporting style How to customize this workflow to your needs Modify AI prompts**: customize the ticket analysis prompt in the "Ticket Summarizer" node to focus on specific aspects like priority, resolution time, or customer impact Adjust ticket filters**: change the JQL queries to filter by status, priority, assignee, or custom date ranges beyond "today" Add more data points**: include additional ticket fields like priority, status, assignee, or custom fields in the analysis Customize email format**: modify the "Format Body" node to change the report structure, add charts, or include additional formatting Set up different schedules**: create multiple versions for different reporting frequencies (hourly, weekly, monthly) Need help customizing? Contact me for consulting and support: 📧 billychartanto@gmail.com
by moosa
Who’s It For This workflow is ideal for HR professionals, recruiters, and small businesses looking to streamline resume screening with AI-powered analysis and CRM integration. What It Does This template automates resume processing by: Capturing resume submissions (name, email, PDF) via JotForm. Converting PDFs to images using PDF.co API. Extracting text with Azure Vision OCR. Analyzing resumes with GPT-4.1 for strengths, improvements, and a score (1–100). Storing submission data in PostgreSQL. Adding high-scoring resumes (>85) to Zoho CRM and sending congratulatory emails. Sending feedback emails for lower-scoring resumes. How to Set Up Configure JotForm Trigger: Add your JotForm API key and form ID (e.g., 252434958811059). Set Up PostgreSQL: Create a resume table with columns: id (SERIAL PRIMARY KEY), given_name (VARCHAR), given_email (VARCHAR), resume_loc (VARCHAR). Add Credentials: Store API keys for PDF.co, Azure Vision OCR, OpenAI, Zoho CRM, and Gmail in n8n’s credential system. Test the Workflow: Submit a test resume via JotForm and verify data flow through each node. Requirements n8n instance (cloud or self-hosted). Accounts with JotForm, PDF.co, Azure Vision, OpenAI, Zoho CRM, and Gmail. PostgreSQL database. How to Customize Adjust the GPT-4.1 prompt for specific job roles. Modify the score threshold (currently 85) in the "if score > 58?" node. Update email templates for personalized messaging. PostgreSQL Table Structure > Node to create table included in workflow.
by Rahul Joshi
📘 Description This workflow automates the employee onboarding process by creating Jira accounts, generating Notion onboarding checklists, crafting AI-generated welcome messages, and sending personalized welcome emails — all automatically. It provides a complete hands-free onboarding experience for HR and IT teams by connecting Jira, Notion, Google Sheets, Gmail, and Azure OpenAI. Failures (like Jira account creation errors) are logged into Google Sheets to ensure full transparency and no missed onboardings. ⚙️ What This Workflow Does (Step-by-Step) 🟢 When Clicking “Execute Workflow” Manually triggers the entire onboarding automation. Useful for testing or initiating onboarding on demand for a new hire. 👤 Define New Hire Profile Data Structures all essential employee information into a clean dataset including name, email, start date, buddy, and access links (Slack, GitHub, Jira, Notion). Acts as the single source of truth for all downstream systems ensuring consistent, error-free onboarding data. 🎫 Create Jira User Account Automatically creates a Jira account for the new employee using REST API calls. Includes email, display name, username, and product access (Jira Software). Removes the need for manual admin setup and ensures immediate access to project boards. ✅ Validate Jira Account Creation Success: Checks if the Jira API response contains a valid accountId. If successful → continues onboarding. If failed → logs error to Google Sheets. Ensures downstream steps don’t continue if Jira setup fails. 📊 Log Jira Provisioning Failures to Error Sheet Appends any account creation errors (duplicate emails, invalid permissions, or API issues) into an “error log sheet” in Google Sheets. Helps HR/IT monitor issues and manually resolve them. Guarantees no silent onboarding failures. 📋 Generate Notion Onboarding Checklist Creates a personalized Notion page titled “{Name} - Onboarding Checklist” that includes: Welcome message Access links (Slack, GitHub, Jira) Assigned buddy details Start date and status Optionally, embedded videos or docs Gives each new hire a structured hub to manage onboarding tasks independently. 🤖 AI-Generated Welcome Message Creator Uses GPT-4o (Azure OpenAI) to craft a friendly, motivational welcome message for the new employee. Incorporates name, buddy, and access details with emojis and warm tone. Ensures every message feels human and engaging — not robotic. 🧠 GPT-4o Language Model Configuration Configures the AI assistant persona for personalized onboarding messages. Ensures tone consistency, friendliness, and empathy across all communications. 🔗 Consolidate Onboarding Data Streams Merges data from Jira, Notion, and AI message generation into a single payload. This ensures the final email contains every onboarding element — access links, checklist URL, and the AI-generated message. 📧 Format Comprehensive Welcome Email Generates a complete HTML-formatted email with: Personalized greeting AI-generated welcome message Clickable links (Jira, Notion, Slack, GitHub) Buddy info and start date Designed for mobile responsiveness and branded presentation. 📬 Send Welcome Email to New Hire Sends the final welcome email to the employee’s inbox with the subject: “Welcome to Techdome, {Name}! 🎉” Includes all essential access information, links, and team introductions — ensuring the new hire starts strong on Day 1. 🧩 Prerequisites Jira Admin API credentials Notion API integration Gmail OAuth2 credentials Azure OpenAI (GPT-4o) access Google Sheets document for logging errors 💡 Key Benefits ✅ Fully automated new hire onboarding ✅ AI-generated personalized communications ✅ Real-time error logging for IT transparency ✅ Seamless integration across Jira, Notion, and Gmail ✅ Professional first-day experience with zero manual work 👥 Perfect For HR teams managing multiple onboardings IT admins automating access provisioning Startups scaling employee onboarding Organizations using Jira + Notion + Gmail stack
by Rahul Joshi
📘 Description: This workflow automates a complete CRM → Sheets → AI → Email reporting pipeline for HighLevel opportunities. It fetches fresh opportunity data from HighLevel, validates and normalizes every record, syncs all structured opportunities into a Google Sheet, merges them into a single dataset, and then uses GPT-4o to generate a clean, Gmail-friendly HTML report summarizing all opportunities for the day. Finally, it emails the formatted report directly to the sales inbox—creating a fully automated, zero-touch Daily Opportunity Insight System. Invalid or incomplete CRM entries are logged automatically, ensuring data hygiene and auditability. ⚙️ What This Workflow Does (Step-by-Step) ▶️ When Clicking ‘Execute Workflow’ (Manual Trigger) Starts the daily reporting pipeline manually or on schedule. 📥 Fetch Opportunities from HighLevel CRM Retrieves the latest opportunities (limit = 5) from HighLevel along with company, contact, source, and pipeline metadata. Acts as the primary CRM input. 🔍 Validate Opportunity Data Payload (IF Node) Checks whether each record contains a valid id. ✅ Valid → proceed to extraction and normalization ❌ Invalid → sent to Google Sheets for cleanup ⚠️ Log Invalid Opportunities to Google Sheets Saves corrupt or incomplete CRM payloads into an error sheet. Supports CRM maintenance and future corrective actions. 🧾 Extract Key Fields from HighLevel Data (Code Node) Pulls only essential fields from each opportunity: id, name, company, email, phone, source, assignedTo, pipelineId, stageId, tags, monetaryValue, and timestamps. Produces a simplified, uniform data structure. 🛠 Normalize Opportunity Structure (Code Node) Cleans and standardizes each opportunity’s schema: ensures consistent field naming, fills contact info when nested, resolves pipeline/stage fields, and finalizes structure for sheet update. 📊 Update Opportunity Records in Google Sheets Upserts (append/update) each opportunity into the ghl database tab of sample_leads_50. Matching key: id Keeps HighLevel CRM and Google Sheets fully synced. 🧩 Merge All Opportunities into a Single JSON Array Combines every normalized opportunity into one array named opportunities. This consolidated payload is passed to GPT-4o for table generation. 🧠 Configure GPT-4o Model (Azure OpenAI) Initializes GPT-4o as the AI engine responsible for generating the final HTML summary. 📄 Generate Daily Opportunity Summary Report (AI Agent) GPT-4o transforms the merged opportunity dataset into a structured HTML report: Daily Opportunity Summary A short descriptive paragraph A full-width Gmail-friendly table with padded cells Header background #f5f5f5 Columns in fixed order: Name, Company, Email, Phone, Source, Pipeline ID, Stage ID, Value, Created At All nulls replaced with “–” Output is pure HTML—no markdown. 📧 Send Daily Opportunity Summary via Gmail Emails the final HTML report to the internal sales inbox with subject: “Daily Opportunity Report – Summary of New Leads” Optimized for Gmail + Outlook rendering. 🧩 Prerequisites HighLevel OAuth connection Azure OpenAI GPT-4o credentials Google Sheets OAuth (Techdome account) Gmail API connection for report delivery 💡 Key Benefits ✅ Automatic syncing of HighLevel CRM opportunities into Sheets ✅ AI-generated HTML dashboards without manual formatting ✅ Clean, readable daily insights for sales teams ✅ Built-in error logging for bad CRM records ✅ Zero manual intervention required after setup 👥 Perfect For Sales & Growth Teams using HighLevel CRM Operations teams maintaining CRM hygiene Agencies needing daily pipeline visibility Organizations wanting automated AI-generated opportunity summaries
by Zeinabsadat Mousavi Amin
This workflow automates the entire UX research planning process — from gathering context to delivering a ready-to-share Google Doc report. Built for UX researchers and designers, it combines AI-powered generation with human feedback loops to make research planning faster, smarter, and more collaborative. 🧠 What it does Collects context** through an online form (organization, product, and research goals) Generates research questions** automatically using an AI Agent Sends approval emails** to the researcher or designer for review and feedback Refines and rewrites** questions based on user input Recommends suitable research methods** for each question, with clear rationales Formats the content** into a structured, professional HTML report Creates and updates a Google Doc** with the final approved research plan 🎯 Who it’s for Perfect for UX teams, design researchers, and product designers who want to streamline their workflow without losing human oversight. Whether you’re preparing a usability study or strategic research plan, this automation helps you focus on insight — not administration. Result: a fully-approved, polished UX Research Plan — ready for collaboration and presentation.
by Apurva Mishra
AI Email Manager: Auto Summary, Labeling, and CRM Logging via n8n + Gemini Overview This workflow turns your Gmail inbox into a fully autonomous AI Email Agent that reads, summarizes, categorizes, and organizes emails in real-time. Built with n8n, Google Gemini, Notion, and Google Sheets, it’s perfect for founders, freelancers, and agencies who receive a ton of emails daily and want to automate the triage process without losing control. How It Works Gmail Trigger — Detects new incoming emails. Process Email Data — Extracts sender info, subject, and content in a clean structured format. AI Email Analyzer — Uses Gemini AI to summarize the email and decide the most relevant label (e.g., Project Updates, Client Requests, Invoices, etc.). Create Gmail Label (if not exists) — Dynamically creates a new label if the AI recommends one that doesn’t exist. AI Agent + Add Label to Email — Applies the correct Gmail label automatically using the message ID. Logs in Notion & Google Sheets — Every processed email (summary, sender, date, label) is logged for tracking and analytics. Who It’s For Entrepreneurs & Founders — Manage investor, client, and product update emails automatically. Agencies & Teams — Classify and track client emails effortlessly across projects. Freelancers & Consultants — Get AI summaries and organize leads without manually labeling emails. Tech Builders — Anyone building AI automation tools and SaaS products around inbox management.
by Rahul Joshi
📘 Description: This end-to-end automation transforms developer support emails into actionable FAQs and sentiment insights using Azure OpenAI GPT-4o, Gmail, Notion, Slack, and Google Sheets. It not only classifies and summarizes each email into a Notion knowledge base but also detects sentiment and urgency, alerts the team on Slack for critical messages, and automatically replies to users with acknowledgment emails. Every failed or malformed payload is transparently logged in Google Sheets — ensuring zero message loss and full visibility into the AI pipeline. The result is a complete AI-driven customer support loop, from inbox to Notion and back to the sender. ⚙️ What This Workflow Does (Step-by-Step) 🟢 Gmail Polling Trigger – Developer Support Inbox Continuously monitors the developer support Gmail inbox every minute for new messages. Extracts the subject, sender, and snippet to initiate AI analysis. 🔍 Validate Email Payload (IF Node) Checks if each incoming email contains valid message data (like message ID and subject). ✅ True Path: continues to AI analysis ❌ False Path: logs error details in Google Sheets for debugging. 🧠 Configure GPT-4o Model (Azure OpenAI) Initializes GPT-4o as the reasoning model for semantic classification of developer support content. 🤖 Analyze & Classify Developer Email (AI Agent) Interprets each email and produces a structured JSON with: Problem summary FAQ category (e.g., API, Billing, UI) 2–3 line solution “Is recurring” flag for common issues. 🧹 Parse & Clean AI JSON Output (Code Node) Removes code formatting (json) and safely parses GPT-4o’s output into clean JSON. If parsing fails, the raw text and error message are sent to Google Sheets for review. 📘 Save FAQ Entry to Notion Database Creates a new FAQ record inside Notion’s “Release Notes” database. Stores the problem, category, and solution as searchable structured fields. 💬 Announce New FAQ in Slack Posts a summary of the new FAQ in Slack with title, category, and answer preview. Includes a link to view the Notion record instantly for team visibility. 🧠 Configure GPT-4o Model (Sentiment Analysis) Sets up another GPT-4o instance focused on understanding tone, emotion, and urgency of each email. ❤️ Analyze Email Sentiment & Urgency (AI Agent) Analyzes the email content to determine: Urgency: Low, Medium, High, Critical Sentiment: Positive, Neutral, Frustrated, Angry Immediate response required? (Yes/No) Provides a short “reason” explaining the classification. 🧹 Parse AI JSON Output – Sentiment Analysis Cleans and validates the JSON from sentiment AI for consistent field names (urgency, sentiment, reason). ⚖️ Filter Critical or High-Urgency Emails (IF Node) Checks if urgency == High or Critical. ✅ True Path: triggers escalation to Slack ❌ False Path: ends quietly to avoid unnecessary noise. 🚨 Alert Team in Slack – Critical Issue Sends an immediate Slack alert with: Email snippet Detected urgency and sentiment Short justification (reason) CTA for urgent action. Ensures fast team response to high-priority issues. 📨 Send Acknowledgment Email to Sender (Gmail Node) Automatically replies to the customer confirming receipt and providing a short AI-generated solution summary. Thanks the user and links the response back to the knowledge base — creating a closed-loop support experience. 🪶 Log Workflow Errors to Google Sheets Appends all failed validations, missing fields, or JSON parsing issues to the “error log sheet.” Provides a live audit trail for monitoring workflow health. 🧩 Prerequisites Gmail account with API access Azure OpenAI (GPT-4o) credentials Notion API integration (for FAQ database) Slack API access (for team alerts) Google Sheets (for logging errors) 💡 Key Benefits ✅ Converts support emails into structured FAQs automatically ✅ Detects sentiment & urgency for faster triage ✅ Keeps Notion knowledge base continuously updated ✅ Sends Slack alerts for critical issues instantly ✅ Maintains transparent error logs in Google Sheets 👥 Perfect For Developer Relations or Product Support Teams SaaS companies managing large support volumes Teams using Gmail, Notion, and Slack for internal comms Startups automating customer response and knowledge creation
by Davide
This workflow automates analyzing Gmail threads and drafting AI-powered replies with the new model Anthropic Sonnet 4.5. This workflow automates the process of analyzing incoming emails and generating context-aware draft replies by examining the entire email thread. Key Advantages ✅ Time-Saving – Automates repetitive email replies, reducing manual workload. ✅ Context-Aware Responses – Replies are generated using the entire email thread, not just the latest message. ✅ Smart Filtering – The classifier prevents unnecessary drafts for spam or promotional emails. ✅ Human-in-the-Loop – Drafts are created instead of being sent immediately, allowing manual review and corrections. ✅ Scalable & Flexible – Can be adapted to different accounts, reply styles, or workflows. ✅ Seamless Gmail Integration – Directly interacts with Gmail threads and drafts via OAuth. How it Works This workflow automates the process of analyzing incoming emails and generating context-aware draft replies by examining the entire email thread. Trigger & Initial Filtering: The workflow is automatically triggered every minute by the Gmail Trigger node, which detects new emails. For each new email, it immediately performs a crucial first step: it uses an AI Email Classifier to analyze the email snippet. The AI determines if the email is a legitimate message that warrants a reply (categorized as "ok") or if it's spam, a newsletter, or an advertisement. This prevents the system from generating replies for unwanted emails. Context Aggregation: If an email is classified as "ok," the workflow fetches the entire conversation thread from Gmail using the threadId. A Code Node then processes all the messages in the thread, structuring them into a consistent format that the AI can easily understand. AI-Powered Draft Generation: The structured conversation history is passed to the Replying email Agent with Sonnet 4.5. This agent, powered by a language model, analyzes the entire thread to understand the context and the latest inquiry. It then drafts a relevant and coherent HTML email reply. The system prompt instructs the AI not to invent information and to use placeholders for any missing details. Draft Creation: The final step takes the AI-generated reply and the original email's metadata (subject, recipient, threadId) and uses them to create a new draft email in Gmail. This draft is automatically placed in the correct email thread, ready for the user to review and send. Set up Steps To implement this automated email reply system, you need to configure the following: Configure Gmail & OpenAI Credentials: Ensure the following credentials are set up in your n8n instance: Gmail OAuth2 Credentials: The workflow uses the same Gmail account for the trigger, fetching threads, and creating drafts. Configure this in the "Gmail Trigger," "Get a thread," and "Create a draft" nodes. OpenAI API Credentials: Required for both the "Email Classifier". Provide your API key in the respective OpenAI Chat Model nodes. Anthropic API Credentials: Required for the main "Replying email Agent." Provide your API key in the respective Antrhopic Chat Model nodes. Review AI Classification & Prompting: Email Filtering: Check the categories in the Email Classifier node. The current setup marks only non-advertising, non-newsletter emails as "ok." You can modify these categories to fit your specific needs and reduce false positives. Reply Agent Instructions: Review the system message in the Replying email Agent. You can customize the AI's persona, tone, and instructions (e.g., making it more formal, or instructing it to sign with a specific name) to better align with your communication style. Need help customizing? Contact me for consulting and support or add me on Linkedin.