by Jimmy Gay
AI-Powered LinkedIn Viral Content Generator & Telegram Bot Disclaimer: This workflow uses community-contributed nodes which are not officially maintained by n8n. Please test thoroughly before running in production. Do not use this template in production without your own independent validation. Overview This workflow empowers you to generate highly viral LinkedIn posts, including both compelling copy and AI-generated custom images, directly from a Telegram chat interface. Leveraging AI-powered research, GPT-based content creation, and community-based integrations, it creates a seamless automation: from prompt to trend analysis, viral copywriting, image generation, and message delivery—all in one flow. Node-by-Node Workflow Explanation Telegram Trigger:** Starts the workflow from a specified Telegram chat by capturing user prompts. Expert Algo (AI Analysis):** Uses OpenAI and Tavily to research current LinkedIn trends, analyzes top-performing content, and produces a content framework plus an AI image prompt. Structured Output Parser:** Validates and formats the AI's returned JSON output. CM Junior:** Generates three LinkedIn post drafts in your own style, based on the viral framework and rules. Structured Output Parser1:** Ensures correct JSON for the drafted posts. Community Manager:** Evaluates the draft posts using additional trend analysis via Tavily, selects the one most likely to go viral. Structured Output Parser2:** Validates the community manager's single post output. Generate Image:** Calls the AI image generator on RapidAPI to create a picture for your post, using the recommended prompt. Split Out/Download Image:** Prepares and downloads the generated image file. Send Telegram Photo and Message:** Sends the chosen LinkedIn post and the generated image to your Telegram bot. Setup Instructions Telegram: Create a bot using @BotFather and get your bot token and target chat ID. OpenAI: Register at OpenAI, obtain your API key. Tavily: Register at Tavily and get your API key. RapidAPI (AI Image Generator): Create an account, subscribe to the "ai-text-to-image-generator-flux-free-api", and copy your API key. Credentials: Use the n8n Credentials Manager to configure each key securely—never hardcode API keys into nodes. Workflow Personalization: Replace all placeholders like Telegram chat ID by referencing the credentials or environment variables. Testing: Run the workflow using your bot and ensure that no personal or sensitive data remains before publishing. Additional Recommendations Update the image generator, Telegram, OpenAI, and Tavily node credentials through the n8n Credentials panel. Custom tailor prompts and output formatting as needed for your LinkedIn content strategy. For security, do not share any personal API keys or chat IDs in your public template.
by Can KURT
n8n – Outlook AI Categorization & Labeling (Fully Automated) > Zero manual mapping. The workflow automatically discovers your Outlook folders, understands the context, assigns the correct category, and moves the email into the right folder. It uses the original Microsoft Outlook nodes plus an AI Agent. You can connect OpenAI or any other LLM provider. ✨ Features Self-Discovery:** Scans your Outlook folders automatically – no manual mapping required. AI-Powered Decisions:** Considers sender, subject, content, links, attachments, timing, and business context. Label + Move:** Assigns the right Outlook category and moves the email into the correct folder. Dual Category Logic:** Can apply both a primary and a secondary category (e.g., Action + Project). Error Handling:** Captures errors and continues without breaking the workflow. Flexible AI Backend:** Replace OpenAI with your own LLM if preferred. 🚀 Setup (5 Steps) Connect Outlook In n8n → Credentials → Microsoft Outlook, grant at least Mail.ReadWrite. Connect AI In n8n → Credentials, set up OpenAI (or another model). Works best with GPT-4.x or GPT-4o. Import the Workflow n8n → Workflows → Import from File/Clipboard and paste the provided JSON. Enable Trigger Adjust the Schedule Trigger (e.g., every 5 minutes). Run & Verify Test run and watch emails get categorized and moved automatically. 🧠 How It Works Schedule Trigger pulls new emails Loop Over Items processes them one by one Markdown / varEmail cleans the content Get Many Folders fetches Outlook categories and folders Summarize + Code prepare category IDs AI Agent applies deep categorization logic Update Category applies the Outlook category Move Folder places the email in the right folder Error Handling ensures workflow stability 🧩 System Prompt Example You are an advanced AI email categorization system. Your mission is to intelligently analyze and categorize emails with maximum accuracy and context awareness. INTELLIGENT CATEGORIZATION ENGINE: Parse all available categories: {{ $json.category }} Multi-layer analysis: Sender, Subject, Body, Links, Attachments Prioritize: Security threats, Action Required, Business Context Specialized: SaaS, Hosting, E-commerce, Finance, Support, Corporate Anti-Spam: Pattern detection, spoofing, red-flag subjects Dual Logic: Primary + Secondary categories when applicable OUTPUT FORMAT (JSON only): { "subject": "EXACT_EMAIL_SUBJECT", "category": "PRIMARY_CATEGORY_FROM_AVAILABLE_LIST", "subCategory": "SECONDARY_CATEGORY_IF_APPLICABLE", "analysis": "Reasoning", "confidence": "HIGH/MEDIUM/LOW" } Available Categories: {{ $json.category }} ⚙️ Parameters & Notes Uses only existing Outlook categories (never invents new ones). Works with any LLM that supports Chat Completions. Requires Mail.ReadWrite permissions. Safe fallback: if unsure, it uses the Action category. 🛡️ Security Processes only what is needed for classification. No external logging of email content unless you configure it. AI provider can be swapped for self-hosted LLMs for compliance. 📄 License & Sharing License:** MIT (or your choice). Tags:** n8n, Outlook, Email, AI, Automation, Categorization Import Method:** Copy/paste workflow JSON into n8n. ✅ Summary Connect → Import → Run. No manual mapping. AI-powered categorization that labels and organizes your Outlook mailbox automatically.
by Shohani
Overview This n8n workflow automatically fetches the latest post from a Telegram channel, translates it using OpenAI, and republishes it to another channel. It supports text, images, and videos. Features Works Without Admin Privileges** - Does not require any bot to be an admin in the source channel. Scheduled execution** - Runs daily at a configurable time AI-powered translation** - Uses OpenAI GPT-4o-mini for natural translations Multi-media support** - Handles text, images, and videos Easy configuration** - All settings in one centralized node Automatic content cleaning** - Removes original channel signatures Prerequisites Required Credentials Telegram Bot API Create a bot via @BotFather Get your bot token Add the bot as an admin to your target channel OpenAI API Sign up at OpenAI Platform Generate an API key Ensure you have sufficient credits Channel Requirements Source Telegram channel must be public Bot must have admin rights in the target channel Setup Instructions 1. Import the Workflow Copy the workflow JSON and import it into your n8n instance The workflow will be imported in inactive state 2. Configure Credentials Telegram Bot Credentials Go to Credentials → Add Credential Select Telegram Enter your bot token from BotFather Test the connection Save as "TelegramBot" OpenAI Credentials Go to Credentials → Add Credential Select OpenAI Enter your OpenAI API key Save as "OpenAI API" 3. Configure Channel Settings Open the "Set Source Channel" node and modify: sourceChannel: "channel_here", // Source channel username (without @) targetChannel: "@your_channel_here", // Target channel (@channel or chat_id) targetLanguage: "Persian", // Target language for translation channelSignature: "signature text" // The channel signature to replaced 4. Adjust Schedule (Optional) Open the "Daily Schedule" node Default: Runs daily at 9:00 AM Modify triggerAtHour and triggerAtMinute as needed 5. Test the Workflow Click "Execute Workflow" to test manually Check if content appears in your target channel Verify translation quality and formatting 6. Activate the Workflow Toggle the workflow to Active status Monitor execution logs for any errors Content Filtering Modify the "Clean Post Content" node to remove specific text patterns: let cleanPost = $input.first().json.post .replaceAll('unwanted_text', '') .replaceAll(/regex_pattern/g, '') .trim(); Multiple Source Channels To monitor multiple channels: Duplicate the workflow Change the sourceChannel in each copy Use different schedules to avoid conflicts Custom Scheduling The Schedule Trigger supports various patterns: Hourly**: { "triggerAtMinute": 0 } Weekly**: { "triggerAtWeekday": 1, "triggerAtHour": 9 } Multiple times**: Use multiple schedule nodes Troubleshooting Common Issues No content fetched Verify source channel is public Check if channel name is correct (without @) Ensure channel has recent posts Translation fails Verify OpenAI API key is valid Check API usage limits and credits Ensure content is not empty Can't send to target channel Verify bot is admin in target channel Check channel username/ID format Test bot permissions manually Compliance Respect copyright and fair use policies Add proper attribution when required Follow Telegram's Terms of Service
by Basil Irfan
WhatsApp RAG Agent (Text + Voice) with Weekly Google Drive Sync One-line summary : Answers WhatsApp in under 100 words, understands voice notes, and retrieves trusted answers from your Google Drive docs (RAG) kept fresh weekly. What this template does Reply to WhatsApp messages* in a polite, human tone with *≤100 words**. Understands text and voice notes**: auto-downloads audio and transcribes to text. Retrieves answers from your knowledge base (RAG)**: Google Drive docs → chunk → embed → store in Supabase → rerank with Cohere. Keeps short-term memory** across the conversation to avoid repetition. Weekly doc sync** from a selected Google Drive folder so non-technical staff can update content without touching n8n. Why it matters Zero busywork:** Update a Google Doc; the bot learns it on the next sync. Trustworthy answers:* Responses come from *your** vetted docs, not random web text. Voice-first friendly:** Handles the WhatsApp reality of “send a voice note.” Safer by design:** Guardrails—no pricing unless in KB, no pushy sales, no medical advice. Triggers WhatsApp Trigger:** Receives incoming messages (text or audio). Google Drive Trigger (weekly):** Detects new/updated files in the chosen folder for ingestion. App credentials required WhatsApp Business Cloud** (App ID, Token, Phone Number ID) OpenAI** (Chat + Embeddings) Cohere** (Rerank) Supabase** (SUPABASE_URL, ANON_KEY) Google Drive** (OAuth2) + target Folder ID Suggested environment variables WHATSAPP_APP_ID= WHATSAPP_TOKEN= WHATSAPP_PHONE_NUMBER_ID= OPENAI_API_KEY= COHERE_API_KEY= SUPABASE_URL= SUPABASE_ANON_KEY= GDRIVE_FOLDER_ID= RAG_TABLE_NAME="documents" # table to store vectors/metadata MAX_ANSWER_WORDS=100 # guardrail for concise replies Architecture overview Answer-time lane (RAG Tool):** Receive WhatsApp message → 2) Transcribe audio if present → 3) Maintain short-term memory → 4) Retrieve from Supabase vectors (topK) → 5) Rerank with Cohere → 6) Compose ≤100-word reply with sources → 7) Send on WhatsApp. Ingest lane (Weekly Sync):** A) Detect Drive file updates → B) Download & convert (Docs → text/plain) → C) Chunk (size/overlap) → D) Embed with OpenAI → E) Upsert to Supabase with metadata & hashes. How it works (node rundown) | # | Node | Key Inputs | Key Outputs | | -- | ------------------------------------------- | ------------------------------- | --------------------------------- | | 1 | WhatsApp Trigger | Incoming message | Raw WhatsApp payload | | 2 | Switch (Attachment presence/type) | Payload | Route: Text or Audio | | 3 | HTTP Request (Audio path) | attachments[0].data_url | Audio file | | 4 | OpenAI – Translate/ASR | Audio file | Transcribed text | | 5 | Merge | Text path + Audio path | Unified text message | | 6 | Simple Memory | Recent turns | Short-term context | | 7 | OpenAI Chat Model | Prompt + message + memory | Draft answer (tool calls allowed) | | 8 | Supabase Vector Tool (retrieve-as-tool) | Query text, topK=10 | Candidate KB passages | | 9 | Cohere Reranker | Candidates | Re-ranked context | | 10 | Send WhatsApp Message | to, body | Reply sent | | 11 | Google Drive Trigger (weekly) | Folder ID, fileUpdated | Changed files | | 12 | Set (File Id) | id from trigger | File ref | | 13 | Google Drive – Download File | Id (Docs→txt) | Raw text | | 14 | Character Text Splitter | chunkSize=2000, overlap=300 | Chunks | | 15 | Default Data Loader | Binary→Document | Clean docs | | 16 | OpenAI Embeddings (ingest) | Chunks | Vectors | | 17 | Supabase Vector Store (insert) | Table: documents | Upserted KB | Notes The KB Tool is the combo of steps 8–9–7 at answer time (retrieve → rerank → answer). The Ingest lane is steps 11–17 (weekly sync of your Drive folder). Setup (7‑minute sprint) Import the workflow JSON. Connect credentials: WhatsApp, OpenAI, Cohere, Supabase, Google Drive. Google Drive Trigger: paste your Folder ID; keep fileUpdated event. Download File: ensure Google Docs convert to text/plain. Supabase Vector Store (insert): set table name to documents (or your schema). Character Text Splitter: keep chunkSize=2000, overlap=300 (balanced recall/latency). Retrieve-as-tool: set topK=10 and enable reranker. Send WhatsApp Message mapping: Recipient: {{$("WhatsApp Trigger").item.json.messages[0].from}} Body: {{$json.output}} Test: Send a text and a voice note to your WhatsApp number → confirm concise answers. Drop a Google Doc into the watched folder → verify it’s chunked/embedded on the next weekly poll (or run ingest nodes once manually). Prompt, tone & guardrails System prompt:** Be polite, human, and concise (≤ MAX_ANSWER_WORDS). Cite or reference only the KB content; if unknown, say so and offer to escalate. No prices unless present in the KB. No medical advice. Temperature:* start at *0.2** for factual replies. Memory window:** keep a short rolling buffer (e.g., last 4–6 turns). Data model (minimum viable) Table documents (example columns): id (uuid) source_url (text) title (text) chunk (text) embedding (vector/float[] depending on extension) chunk_hash (text) updated_at (timestamp) Indexes Unique index on chunk_hash to dedupe Index on updated_at for syncs Observability & ops Log question, selected chunk ids/hashes, and final response to a DB/Sheet for QA. Add a low-confidence route (score threshold) → Slack/Telegram escalation to a human. Track latency and token usage to tune topK and chunk sizes. Customization Latency vs quality:** try topK=6–8 and chunkSize=1200 for speed. Languages:** ASR node can be swapped for native multilingual output. Escalation:** add channel handoff on low confidence or no KB hits. Sync cadence:* change Drive Trigger to *daily** if content updates frequently. Safety & compliance No medical advice.** If uncertain or clinical, ask to schedule a consult or refer to the right department. PII:** Don’t log full phone numbers in plaintext analytics; hash where possible. Prices & rates:** Only answer if present in the KB; otherwise hand off to front desk. Troubleshooting No reply sent:* Ensure *Send message** node reads {{$json.output}} (Agent’s response property). Audio path failing:** Confirm attachments[0].data_url exists and HTTP node fetches a valid file. KB not updating:** Manually execute the ingest lane; check rows in Supabase. Irrelevant answers:* Lower temperature to *0.2, increase overlap to **400, and verify Drive docs are clean and structured. Categories & tags Categories:** AI, Customer Support, Healthcare Ops, RAG, WhatsApp Tags:** WhatsApp, RAG, Google Drive, Supabase, OpenAI, Cohere, Voice Notes Pricing (rough, BYO keys) n8n:** self-host free; n8n.cloud billed by plan. OpenAI, Cohere:** usage-based by tokens/calls. Supabase:** free tier + usage; vector storage billed by size. WhatsApp Cloud:** Meta pricing per conversation. Nodes used in workflow WhatsApp Trigger, Switch, HTTP Request, OpenAI (ASR + Chat + Embeddings), Merge, Simple Memory, Supabase Vector Tool (retrieve), Cohere Reranker, Google Drive Trigger, Set, Google Drive – Download, Character Text Splitter, Default Data Loader.
by Yaron Been
Create Multi-Channel Content with O3 Director & GPT-4 Specialist Agents This n8n workflow creates a complete AI-powered content department. It starts when a chat request is received, then a Content Director Agent (powered by OpenAI O3) analyzes the request and delegates tasks to specialized agents (blogs, social, video, email, website, strategy). Each agent is powered by GPT-4.1-mini, keeping costs low and quality high. ✅ 📩 Section 1: Trigger & Director Setup ⚙️ Nodes 1️⃣ When Chat Message Received What it does:** Starts the workflow whenever a user sends a content request. Why it’s useful:** Allows real-time or on-demand content creation from chat inputs. 2️⃣ Content Director Agent (O3) What it does:** Analyzes user request, defines the best content mix, and delegates tasks to specialist agents. Why it’s useful:** Keeps your brand voice consistent and ensures all channels align to a unified content strategy. 💡 Beginner Benefit ✅ Single entry point → just type your content idea once ✅ AI Director coordinates everything for you ✅ No need to manage multiple tools ✅ 👥 Section 2: Specialist Content Agents Each request gets routed to one (or several) of these agents, depending on the strategy. 3️⃣ Blog Content Writer Long-form articles, editorials, and thought leadership pieces. 4️⃣ Social Media Content Creator Social posts, captions, hashtags, and community content. 5️⃣ Video Script Writer YouTube scripts, explainer videos, and video marketing content. 6️⃣ Email Newsletter Writer Campaigns, nurture sequences, and newsletter copy. 7️⃣ Website Copy Specialist Landing pages, product descriptions, and conversion-focused web copy. 8️⃣ Content Strategist & Planner Editorial calendars, campaign planning, and audience strategy. 💡 Beginner Benefit ✅ Each agent is an expert in its field ✅ Powered by GPT-4.1-mini → faster and cheaper ✅ Parallel execution → all content types can be generated at once ✅ 🧠 Section 3: Language Models & Execution Flow O3 Model → Content Director** Handles analysis, strategy, and delegation. GPT-4.1-mini → All Specialists** Powers blog, social, video, email, website, and strategy agents. Think Node** Helps the Content Director organize reasoning before delegating tasks. 💡 Beginner Benefit ✅ AI Director (O3) = smart leadership ✅ Specialists (GPT-4.1-mini) = cost-efficient execution ✅ Built-in reasoning = better, more aligned campaigns 📊 Workflow Overview | Section | What Happens | Key Benefit | | --------------------------- | --------------------------------------------------------------- | -------------------------- | | 📩 Trigger & Director Setup | Workflow starts from chat → Content Director interprets request | Centralized control | | 👥 Specialist Agents | Each AI agent produces tailored content | Multi-channel coverage | | 🧠 Models & Flow | O3 for Director, GPT-4.1-mini for specialists | Cost-efficient + strategic | 📌 How You Benefit Overall ✅ One input → full content campaign ✅ Consistent brand voice across all platforms ✅ Cost-effective (O3 only for strategy, GPT-4.1-mini for bulk work) ✅ Ready-to-publish content in minutes ✨ You’ve basically built an AI marketing department inside n8n — no extra staff required! 🚀
by Aditya Malur
This n8n template helps you turn business cards into qualified sales opportunities — instantly. No more lost leads after events or networking meetups. Just send a business card photo via Telegram, and let AI handle the rest. The workflow extracts details (name, company, role, email, phone) using Google Vision OCR, analyzes context with OpenAI, and then generates personalized WhatsApp or email messages to help you follow up faster. Use Cases Capture leads instantly from events or meetups. Auto-analyze business card data into structured CRM entries. Send hyper-personalized WhatsApp messages or emails within minutes. Ideal for founders, marketers, and business development teams. How it Works Trigger: Send a photo of a business card via Telegram. Extract: Google Vision OCR reads and extracts all text from the image. Process: The extracted text is sent to an AI agent (OpenAI GPT-4.1-mini) for: Contact parsing (name, company, role, email, phone) Business/industry inference AI-generated personalized follow-up messages Output: The system returns structured JSON containing: Email subject & body WhatsApp draft message Fit score, opportunities, and next action steps How to Use Import the JSON template into n8n. Add your Telegram Bot Token to the “Telegram Trigger” node. Add your Google Cloud Vision API key in the HTTP Request node. Add your OpenAI credentials in the “OpenAI Chat Model” node. (Optional) Connect the output to Google Sheets, Airtable, or your CRM. Run the workflow — take a photo of a business card and watch the magic happen! Requirements Telegram Bot for image input Google Vision API Key OpenAI API Key (Optional) Integration with CRM or WhatsApp API Customization Ideas Replace Telegram with Email or WhatsApp triggers. Push qualified leads directly into HubSpot or Notion. Automate multi-step sequences based on AI fit scores.
by Samir Saci
Tags: Logistics, Supply Chain, Warehouse Operations, Paperless processes, Quality Management Context Hi! I’m Samir — Supply Chain Engineer, Data Scientist based in Paris, and founder of LogiGreen. > Let us use n8n to help small companies digitalise their logistics and supply chain! This workflow helps warehouse operators generate a complete damage report without needing to write anything manually. In warehouse operations, damaged pallets must be reported quickly and consistently. You can automate the entire process using AI to analyse photos of the damages. 📬 For business inquiries, you can find me on LinkedIn Example of damage report The process starts with instructions sent with the operator: A photo of the damaged pallets is shared with the bot: A complete report is generated and sent by email: 🎥 Tutorial A complete tutorial (with explanations of every node) is available on YouTube: Who is this template for? This template is ideal for companies with limited IT ressources: Warehouse operators** who need a fast reporting tool Quality teams** who want consistent and structured reports 3PLs and logistics providers** looking to digitalise damage claims Manufacturers and retailers** with high inbound pallet volumes Anyone using Telegram** on the warehouse floor for quick interactions What does this workflow do? This workflow acts as an AI-powered damaged goods reporting assistant using Telegram, OpenAI Vision and Gmail. A operator sends a picture of the damaged pallet via Telegram. The workflow downloads the image and sends it to GPT-4o for damage analysis. The bot replies and asks for a photo of the pallet barcode. The barcode picture is processed by GPT-4o Mini to extract the pallet number. The workflow combines both results (damage analysis + pallet ID). It generates an HTML email report with: damage summary, observed issues, severity level and recommended actions The report is automatically sent via Gmail to the configured recipient. The operator receives a confirmation message in Telegram. The processes does not require any data input form the operator, only to take pictures! Next Steps Before running the workflow, follow the sticky notes and configure: Connect your Telegram Bot API Add your OpenAI API Key in the AI nodes Connect your Gmail credentials Update the recipient email in the “Send Report by Email” node Submitted: 20 November 2025 Template designed with n8n version 1.116.2
by minh
Who’s it for This template is designed for anyone who wants to use Telegram as a personal AI assistant hub. If you often juggle tasks, emails, calendars, and expenses across multiple tools, this workflow consolidates everything into one seamless AI-powered agent. What it does Jarvis listens to your Telegram messages (text or audio) and processes them with OpenAI. Based on your request, it can:
by Iternal Technologies
This workflow contains community nodes that are only compatible with the self-hosted version of n8n. Blockify® Data Optimization Workflow Blockify Optimizes Data for RAG - Giving Structure to Unstructured Data for ~78X Accuracy, when pairing Blockify Ingest and Blockify Distill Learn more at https://iternal.ai/blockify Get Free Demo API Access here: https://api.blockify.ai/register Read the Technical Whitepaper here: https://iternal.ai/blockify-results See example Accuracy Comparison here: https://iternal.ai/case-studies/medical-accuracy/ Blockify is a data optimization tool that takes messy, unstructured text, like hundreds of sales‑meeting transcripts or long proposals, and intelligently optimizes the data into small, easy‑to‑understand "IdeaBlocks." Each IdeaBlock is just a couple of sentences in length that capture one clear idea, plus a built‑in contextualized question and answer. With this approach, Blockify improves accuracy of LLMs (Large Language Models) by an average aggregate 78X, while shrinking the original mountain of text to about 2.5% of its size while keeping (and even improving) the important information. When Blockify's IdeaBlocks are compared with the usual method of breaking text into equal‑sized chunks, the results are dramatic. Answers pulled from the distilled IdeaBlocks are roughly 40X more accurate, and user searches return the right information about 52% more accurate. In short, Blockify lets you store less data, spend less on computing, and still get better answers- turning huge documents into a concise, high‑quality knowledge base that anyone can search quickly. Blockify works by processing chunks of text to create structured data from an unstructured data source. Blockify® replaces the traditional "dump‑and‑chunk" approach with an end‑to‑end pipeline that cleans and organizes content before it ever hits a vector store. Admins first define who should see what, then the system ingests any file type—Word, PDF, slides, images—inside public cloud, private cloud, or on‑prem. A context‑aware splitter finds natural breaks, and a series of specially developed Blockify LLM model turns each segment into a draft IdeaBlock. GenAI systems fed with this curated data return sharper answers, hallucinate far less, and comply with security policies out of the box. The result: higher trust, lower operating cost, and a clear path to enterprise‑scale RAG without the cleanup headaches that stall most AI rollouts.
by Carl Danley
Overview This n8n template demonstrates how you can generate an AI-produced weather analysis of your local radar loop and home assistant precipitation sensor(s) to keep your family informed of National Weather Service Alerts. With as crazy as things have been lately in the open world, how will you and your family know when a severe or extreme alert impacts your area? How it Works This workflow is triggered by a webhook which takes a latitude and longitude json payload to identify the area for monitoring. Then, it fetches the National Weather Service Alerts and filters them down to alerts which are currently active and their severity. Next, it fetches the local precipitation value from your Home Assistant instance (a value like "Light Rain" or "No Rain", etc) coupled with your respective weather.gov radar loop image. It then submits this data to OpenAI and produces an output regarding the image analysis. Finally, it takes this analysis and uses OpenAI to again generate a short summary. How to Use Import the workflow into your n8n instance Update the credentials in the problematic nodes Make sure you adjust the radar loop image that is being used Requirements A Home Assistant Instance (you could remove this data if you wanted) An OpenAI account for LLM and image analysis
by Alvin Chandra
AI Creative Director: Transform Phone Photos into Viral Ad Campaigns Description Turn amateur product snapshots into high-conversion advertising assets automatically. This agentic workflow acts as your personal Art Director. Simply send a photo to a Telegram bot, and the AI will analyze the image, conceptualize 3 distinct marketing angles, and generate professional-grade commercial photography. How it Works Ingest: Receives a photo and user caption via Telegram. Analyze: Uses Gemini Vision to understand the product and critique the original image quality. Ideate: A LangChain Agent (powered by Claude Sonnet via OpenRouter) acts as a Creative Director to brainstorm 3 unique ad concepts (e.g., Cyberpunk, Luxury, Minimalist). Create: Generates high-fidelity images using the Gemini 3 Pro API. Deliver: Sends the new ad creatives back to you on Telegram. Key Features Multi-Modal Pipeline:** Chains Vision models (Gemini) with Reasoning models (Claude) for superior context awareness. Structured Output:** Uses LangChain's Structured Output Parser to ensure consistent JSON formats. Secure Configuration:** Features a centralized CONFIG node to easily manage API keys and user IDs without editing complex logic. Visual Verification:* Includes an authentication check to ensure only *you (the authorized user) can trigger the bot. Prerequisites n8n Version:** 1.0+ (requires LangChain nodes). Telegram:** A Bot Token and your personal User ID. Google Cloud:** An API Key with access to Gemini models. OpenRouter:** An API Key for accessing Claude Sonnet (or your preferred LLM). Setup Instructions Import the Workflow: Paste the JSON into your n8n canvas. Open the CONFIG Node: Locate the green node at the start of the workflow. Enter Credentials: authorized_user_id: Your numeric Telegram User ID (to prevent strangers from using your API credits). telegram_bot_token: Your Bot Father token. google_api_key: Your Gemini API key. Activate: Toggle the workflow to "Active" and send a photo to your bot! Tags #AI #LangChain #Gemini #Telegram #Marketing #Agent #ImageGeneration #OpenRouter
by Anton Bezman
Dungeons and Goblins — AI Telegram Voice Adventure with Persistent Memory This n8n template demonstrates how to use an AI agent with persistent memory to run a structured, rules-driven fantasy role-playing game entirely through Telegram voice messages. The workflow acts as a Dungeon Master, narrating scenes, resolving mechanics, performing dice rolls when authorized, and explicitly saving game state between turns. How it works A player actions are provided to Telegram bot via voice messages. The AI agent loads the current game state from n8n memory. A strict system prompt enforces rules, turn flow, and narration. When an action requires a dice roll, the agent waits for player authorization. Once authorized, the AI rolls, resolves the outcome, and applies changes. All state updates are emitted as structured data and saved to memory. The request and response are processed in Groq's STT and TTS. Use cases Solo text-based fantasy campaigns Persistent AI-driven adventures Testing stateful AI agents in n8n Educational examples of memory-aware workflows Requirements Groq API token (free tier supported) Telegram bot API token