Harsh Maniya
Crafting Intelligent AI Solutions | AI Engineer building the next generation of intelligent workflows on n8n. Automating the complex, one node at a time.
Templates by Harsh Maniya
E-commerce assistant for Shopify & WooCommerce with GPT-4o, Gemini & RAG
🤖 Universal E-Commerce AI Assistant (Shopify, WooCommerce & RAG) This powerful n8n workflow deploys a sophisticated, multi-talented AI chatbot designed to streamline your e-commerce and customer support operations. The AI assistant can intelligently understand user queries and route them to the correct specialized agent, whether it's for Shopify, WooCommerce, or general knowledge questions answered by a Retrieval-Augmented Generation (RAG) system. This template automates responses to a wide range of inquiries, from checking Shopify order statuses with GraphQL to fetching product lists from WooCommerce, and even answering general questions by looking up information in a Pinecone vector database. How It Works ⚙️ The workflow operates in a series of logical steps, starting from the moment a user sends a message. 💬 Chat Trigger: The workflow activates when a user sends a message in the n8n chat interface. It captures the user's input and a unique session ID to track the conversation. 🧠 Intelligent Routing: The user's query is first sent to a Router Agent powered by GPT-4o-mini. This agent's sole purpose is to classify the intent of the message and output one of three keywords: SHOPIFY, WOOCOMMERCE, or None of them. 🔀 Conditional Branching: Based on the Router's output, a series of IF nodes direct the conversation down one of three paths: General Queries Path Shopify Path WooCommerce Path 📚 General Queries (RAG): If the query is not about e-commerce, it's handled by a RAG agent. Embedding: The user's question is converted into a vector embedding using AWS Bedrock. Retrieval: The workflow searches a Pinecone Vector Store to find the most relevant information from your knowledge base. Generation: A GPT-4o-mini agent receives the context from Pinecone and generates a comprehensive, helpful answer. 🛍️ E-Commerce Specialists: If the query is about Shopify or WooCommerce, it's passed to a dedicated agent. Shopify Agent: This agent uses Google Gemini and has a suite of tools to manage Shopify tasks. It can Get Order info, Fetch All Products, or run complex queries using the powerful GraphQL tool. WooCommerce Agent: This agent also uses Google Gemini and is equipped with tools to Fetch Order Details and Fetch All Products from a WooCommerce store. 🗣️ Conversation Memory: Each agent (Router, General, Shopify, WooCommerce) is connected to its own Memory node. This allows the chatbot to remember previous parts of the conversation for a more natural and context-aware interaction. 🏁 Merge & Respond: All three paths converge at a final Merge node. This ensures that no matter which agent handled the request, the final answer is streamlined into a single output and sent back to the user in the chat. Nodes Used 🔗 Triggers: Chat Trigger: Starts the workflow when a chat message is received. AI & Agents: AI Agent: Four separate agents for Routing, Shopify, WooCommerce, and General Queries. OpenAI Chat Model: Uses GPT-4o-mini for the Router and General Queries agent. Google Gemini Chat Model: Uses Google Gemini for the Shopify and WooCommerce agents. Tools & Data: Shopify Tool: To get products and order information from Shopify. WooCommerce Tool: To get products and order information from WooCommerce. GraphQL Tool: For advanced, custom queries to the Shopify API. Pinecone Vector Store: To retrieve context for the RAG agent. AWS Bedrock Embeddings: To create vector embeddings for Pinecone. Logic & Memory: IF Node: To conditionally route the workflow. Merge Node: To consolidate the different branches before ending. Window Buffer Memory: Four nodes to provide conversational memory to each agent. Setup Guide 🛠️ To use this workflow, you'll need to configure several nodes with your own credentials and settings. 1\. AI Model Credentials OpenAI: Create an API key in your OpenAI Platform dashboard. Add this credential to the Router Model and GPT-4o-mini nodes. Google Gemini: Create an API key in your Google AI Studio dashboard. Add this credential to the Shopify Chat Model and WooCommerce Chat Model nodes. 2\. E-Commerce Platform Credentials Shopify: You will need a Shopify Access Token. Follow the n8n documentation to generate one. Add the credential to the Fetch All Products and Get Order info nodes. WooCommerce: Create API credentials from your WordPress dashboard. Add the credential to the Fetch All Products2 and Fetch Order Details nodes. 3\. RAG System Credentials (Pinecone & AWS) Pinecone: Sign up for a Pinecone account and create an API key. Add your Pinecone credentials in n8n. In the Pinecone Vector Store node, set the pineconeIndex to the name of your index. You must have a pre-existing index with data for the RAG to work. AWS: Create an AWS account and an IAM user with programmatic access to Amazon Bedrock. Add your AWS credentials in n8n. Select your AWS credentials in the AWS Bedrock Embeddings node. 4\. GraphQL Node Configuration In the GraphQL node, you must update the endpoint URL. Replace the placeholder https://{subdomain}.myshopify.com/admin/api/2025-04/graphql.json with your own Shopify store's GraphQL API endpoint.
By Harsh ManiyaDaily AI news summaries with Gemini 2.5 Flash to Telegram
Get Daily AI News Briefings in Telegram with Gemini & n8n Tired of sifting through endless articles to stay updated on Artificial Intelligence? This workflow transforms you into an intelligence analyst by automatically fetching the latest AI news, using Google's Gemini to create a professional summary, and delivering a polished briefing directly to your Telegram chat. Stop chasing news and let the insights come to you\! 🚀 🚀 What this workflow does 📰 Monitors an RSS Feed: Automatically triggers whenever a new article is published on the specified AI news feed (artificialintelligence-news.com). 🧠 Scrapes Article Content: Uses Jina AI to read and extract the full, clean text from the article link. 🤖 Generates an AI Report: Leverages the power of the Google Gemini 2.5 Flash model to analyze the article and generate a structured, insightful intelligence briefing based on a specialized prompt. 📨 Delivers to Telegram: Sends the final, perfectly formatted markdown report as a message to your chosen Telegram chat. 🛠️ What you'll need to get started An n8n instance (cloud or self-hosted). Jina AI Credentials: You will need an API key from Jina AI. You can get one from the Jina AI website. Google Gemini Credentials: You'll need an API key for the Gemini model. You can generate one for free from Google AI Studio. Telegram Bot Credentials: You will need a Telegram Bot Token and your destination Chat ID. You can learn how to get these from the n8n Telegram documentation. ⚙️ How to use this workflow Copy the workflow: Click "Use template" to import this workflow into your n8n canvas. Configure Jina AI Node: Select the "Read News from AI News Website" node. In the "Credentials" section, add your Jina AI API key. Configure Gemini Node: Select the "Gemini 2.5 Flash" node on the canvas. In the "Credentials" section, add your Google Gemini API key. Configure Telegram Node: Select the "Send a text message" node. Add your Telegram API credentials. In the "Chat ID" field, enter the specific chat, group, or channel ID where you want to receive the news briefings. Activate the workflow: Hit the "Activate" toggle in the top right corner. Your automated AI analyst is now live\! ✨ Bonus: Expand your intelligence briefing The prompt in the "Generate Report" node is designed to synthesize information from multiple sources. You can easily expand this workflow to create even more comprehensive reports: Add another RSS Feed: Copy the "AI-News Feed" node and change the Feed URL to another source. Connect the Nodes: Drag the output of the new RSS feed node to the "Read News from AI News Website" node. That's it\! The workflow will now process articles from both feeds and the Gemini node will automatically identify common themes and unique stories as instructed in the prompt.
By Harsh ManiyaCreate stunning AI images directly from WhatsApp with Gemini
📱🤖 Create Stunning AI Images Directly from WhatsApp with Gemini This workflow transforms your WhatsApp into a personal AI image generation studio. Simply send a text message with your idea, and this bot will use the advanced prompt engineering capabilities of Gemini 2.5 Pro to craft a detailed, artistic prompt. It then uses Gemini 2.0 Flash to generate a high-quality image from that prompt and sends it right back to your chat. It's a powerful yet simple way to bring your creative ideas to life, all from the convenience of your favorite messaging app\! What this workflow does Listens for WhatsApp Messages: The workflow starts automatically when you send a message to your connected WhatsApp number. Enhances Your Idea with AI: It takes your basic text (e.g., "a knight on a horse") and uses Gemini 2.5 Pro to expand it into a rich, detailed prompt perfect for image generation (e.g., "A cinematic, full-body shot of a stoic knight in gleaming, ornate silver armor, riding a powerful black warhorse through a misty, ancient forest. The scene is lit by ethereal morning sunbeams piercing the canopy, creating dramatic volumetric lighting and long shadows. Photorealistic, 8K, ultra-detailed, award-winning fantasy concept art."). Generates a Unique Image: It sends this enhanced prompt to the Google Gemini 2.0 Flash image generation API. Prepares the Image: The API returns the image in Base64 format, and the workflow instantly converts it into a binary file. Sends it Back to You: The final, high-quality image is sent directly back to you in the same WhatsApp chat. Nodes Used 🟢 WhatsApp Trigger: The entry point that listens for incoming messages. 🧠 LangChain Chain (LLM): Uses Gemini 2.5 Pro for advanced prompt engineering. ➡️ HTTP Request: Calls the Google Gemini 2.0 Flash API to generate the image. 🔄 Convert to File: Converts the Base64 image data into a sendable file format. 💬 WhatsApp: Sends the final image back to the user. Prerequisites To use this workflow, you will need: An n8n instance. A WhatsApp Business Account connected to n8n. You can find instructions on how to set this up in the n8n docs. A Google Gemini API Key. You can get one for free from Google AI Studio. How to use this workflow Get your Google Gemini API Key: Visit the Google AI Studio and create a new API key. Configure the Gemini 2.5 Pro Node: In the n8n workflow, select the Gemini 2.5 Pro node. Under 'Connect your account', click 'Create New' to add a new credential. Paste your Gemini API key from the previous step and save. Configure the Generate Image (HTTP Request) Node: Select the Generate Image node. In the properties panel on the right, find the Query Parameters section. In the 'Value' field for the key parameter, replace "Your API Key" with your actual Google Gemini API Key. Connect WhatsApp: Select the WhatsApp Trigger node. Follow the instructions to connect your WhatsApp Business Account credential. If you haven't created one, the node will guide you through the process. Activate and Test: Save the workflow using the button at the top right. Activate the workflow using the toggle switch. Send a message to your connected WhatsApp number (e.g., "A futuristic city in the style of Van Gogh"). The bot will process your request and send a stunning AI-generated image right back to you\!
By Harsh ManiyaResearch assistant for WhatsApp using Twilio, Perplexity and Claude
Build an AI Research Assistant for WhatsApp with Perplexity and Claude 💡 Ever wished you could get a deep, multi-source research report on any topic, delivered directly to your WhatsApp chat in seconds? This workflow transforms your WhatsApp into a powerful, on-demand research assistant, perfect for students, professionals, and curious minds. Leveraging the deep research capabilities of Perplexity, the nuanced formatting skills of Anthropic's Claude, and the messaging power of Twilio, this workflow automates the entire process from query to polished answer. Ask it anything, and receive a well-structured, easy-to-read summary moments later. What This Workflow Does 🚀 📲 Listens for Incoming Queries: The workflow starts when a user sends a message to your configured Twilio WhatsApp number. 🧠 Performs Deep Research: It takes the user's query and feeds it to the Perplexity node, which uses the sonar-pro model to conduct a comprehensive, multi-source analysis on the topic. 🎨 Polishes the Content: The raw, detailed research report from Perplexity is then passed to an Anthropic Claude model. This crucial step refines the text, adds engaging emojis, ensures it's under the WhatsApp character limit, and formats it perfectly for mobile viewing. 💬 Sends an Instant Reply: The final, beautifully formatted summary is sent directly back to the user's WhatsApp chat via Twilio, completing the entire request. Nodes Used 🛠️ Webhook: To receive the initial message from Twilio and trigger the workflow. Perplexity: To perform the AI-powered deep research on the user's query. Anthropic (via LangChain): To connect to and use the Claude model for reformatting and polishing the content. Twilio: To send the final, formatted message back to the user on WhatsApp. How to Set Up This Workflow ⚙️ This workflow requires careful setup between n8n and Twilio to function correctly. Follow these steps closely. 1\. Prerequisites ✅ You will need accounts for the following services: n8n (Cloud or self-hosted) Twilio Perplexity AI (for an API key) Anthropic (for a Claude API key) 2\. Configure Credentials 🔑 In your n8n instance, add your API keys for Twilio, Perplexity, and Anthropic. You can add credentials in n8n by going to the Credentials tab in the left-hand menu. Learn more about managing credentials in n8n. 3\. Set Up Your Twilio WhatsApp Number 📱 Log in to your Twilio account. Either purchase a phone number that is WhatsApp-enabled or use the free Twilio Sandbox for WhatsApp for testing. Follow Twilio's guide to connect your number or sandbox to the WhatsApp Business API. Read Twilio's Getting Started with WhatsApp Guide. 4\. Expose Your n8n Webhook URL 🌐 For Twilio to communicate with n8n, your n8n webhook URL must be publicly accessible. Open the Fetch Whatsapp Request (Webhook) node in the workflow. You will see two URLs: Test and Production. For this guide, we will use the Test URL. If you are running n8n locally or on a private server, you must expose this URL to the public internet. You can do this easily using n8n's built-in tunnel feature. Start n8n from your computer's command line with the following command: bash n8n start --tunnel After starting, n8n will provide you with a public "tunnel URL". It will look something like https://[subdomain].hooks.n8n.cloud/. Copy this public tunnel URL. Learn more about tunneling in the n8n docs. 5\. Connect Twilio to Your n8n Webhook 🔗 In your Twilio Console, navigate to the settings for the phone number (or sandbox) you configured in Step 3. Scroll down to the Messaging section. Find the field labeled "A MESSAGE COMES IN". Paste your n8n Test URL (or your public tunnel URL from the previous step) into this field. Ensure the dropdown next to it is set to HTTP POST. Click Save. 6\. Activate and Test Your Workflow ▶️ Go back to your n8n canvas. Click "Test workflow" in the top right corner. This will put your webhook in a "listening" state. Now, send a message from your personal WhatsApp to your configured Twilio WhatsApp number. You should see the workflow execute successfully in n8n and receive a formatted reply on WhatsApp\! Once you've confirmed it works, save and activate the workflow to have it run permanently. Further Reading 📚 n8n Webhook Node Documentation n8n Twilio Node Documentation Perplexity API Documentation Anthropic API Documentation
By Harsh ManiyaWhatsApp fact-checking bot with Perplexity AI and Twilio
✅💬Build Your Own WhatsApp Fact-Checking Bot with AI Tired of misinformation spreading on WhatsApp? 🤨 This workflow transforms your n8n instance into a powerful, automated fact-checking bot\! Send any news, claim, or question to a designated WhatsApp number, and this bot will use AI to research it, provide a verdict, and send back a summary with direct source links. Fight fake news with the power of automation and AI\! 🚀 How it works ⚙️ This workflow uses a simple but powerful three-step process: 📬 WhatsApp Gateway (Webhook node): This is the front door. The workflow starts when the Webhook node receives an incoming message from a user via a Twilio WhatsApp number. 🕵️ The Digital Detective (Perplexity node): The user's message is sent to the Perplexity node. Here, a powerful AI model, instructed by a custom system prompt, analyzes the claim, scours the web for reliable information, and generates a verdict (e.g., ✅ Likely True, ❌ Likely False). 📲 WhatsApp Reply (Twilio node): The final, formatted response, complete with the verdict, a simple summary, and source citations, is sent back to the original user via the Twilio node. Setup Guide 🛠️ Follow these steps carefully to get your fact-checking bot up and running. Prerequisites A Twilio Account with an active phone number or access to the WhatsApp Sandbox. A Perplexity AI Account to get an API key. 1\. Configure Credentials You'll need to add API keys for both Perplexity and Twilio to your n8n instance. Perplexity AI: Go to your Perplexity AI API Settings. Generate and copy your API Key. In n8n, go to Credentials \& New, search for "Perplexity," and add your key. Twilio: Go to your Twilio Console Dashboard. Find and copy your Account SID and Auth Token. In n8n, go to Credentials \& New, search for "Twilio," and add your credentials. 2\. Set Up the Webhook and Tunnel To allow Twilio's cloud service to communicate with your n8n instance, you need a public URL. The n8n tunnel is perfect for this. Start the n8n Tunnel: If you are running n8n locally, you'll need to expose it to the web. Open your terminal and run: bash n8n start --tunnel Copy Your Webhook URL: Once the tunnel is active, open your n8n workflow. In the Receive Whatsapp Messages (Webhook) node, you will see two URLs: Test and Production. Copy the Test/Production URL. This is the public URL that Twilio will use. 3\. Configure Your Twilio WhatsApp Sandbox Go to the Twilio Console and navigate to Messaging \& Try it out \& Send a WhatsApp message. Select the Sandbox Settings tab. In the section "WHEN A MESSAGE COMES IN," paste your n8n Production Webhook URL. Make sure the method is set to HTTP POST. Click Save. How to Use Your Bot 🚀 Activate the Sandbox: To start, you (and any other users) must send a WhatsApp message with the join code (e.g., join given-word) to your Twilio Sandbox number. Twilio provides this phrase on the same Sandbox page. Fact-Check Away\! Once joined, simply send any claim or question to the Twilio number. For example: Did Elon Musk discover a new planet? Within moments, the workflow will trigger, and you'll receive a formatted reply with the verdict and sources right in your chat\! Further Reading & Resources 🔗 n8n Tunnel Documentation Twilio for WhatsApp Quickstart Perplexity AI API Documentation
By Harsh ManiyaAutomate daily Hindu festival posts on X with Google Gemini and GPT-4o Mini 🤖
✨ Automate Daily Hindu Festival Posts on X (Twitter) with AI 🐦 This workflow automates the entire process of creating and publishing culturally rich social media content about Hindu festivals. It starts by building a comprehensive festival calendar for the year in a Google Sheet, then runs daily to post engaging, bilingual updates on X (formerly Twitter). 🗓️ The workflow uses a sophisticated dual-AI system: Google Gemini acts as a content generator creating multiple post options ✍️, while OpenAI's GPT-4o Mini acts as a discerning social media manager, selecting the very best post for publication. 🧠 This ensures your content is not only automated but also high-quality and optimized for engagement. How it works ⚙️ This workflow operates in two distinct stages: Part 1: Data Population (One-Time Manual Run) 🔍 Fetch Festival Data: Manually trigger the workflow to scrape a list of 2025 Hindu festivals from a public calendar using the Jina AI Reader API. ✨ Enrich with AI: For each festival, a Google Gemini model researches and extracts key details: The festival's name in Hindi. A concise description of its significance in English. A Hindi translation of the description. 📝 Store in Google Sheets: The enriched data for the entire year is then systematically organized and saved in a designated Google Sheet, creating a content calendar. Part 2: Daily Automated Posting ⏰ Daily Trigger: A Schedule Trigger node activates the workflow every morning at 8 AM. ✔️ Check for Festivals: The workflow gets today's date and checks the Google Sheet to see if there is a corresponding festival. 🎨 Generate Post Options: If a festival is scheduled for the day, Google Gemini generates three distinct and engaging post options for X. Each post is crafted to be concise, use a mix of English and Hindi, and include relevant emojis and hashtags. 🏆 Select the Best Post: OpenAI's GPT-4o Mini then evaluates the three generated posts based on criteria like clarity, engagement potential, and effective use of language. It selects the single most impactful post. 🚀 Publish to X: The winning post is automatically published to your connected X account. Features ⭐ 🤖 Fully Automated Content Pipeline: From data collection to final publication, no manual intervention is needed after the initial setup. 🧠 Dual-AI System: Leverages Google Gemini for creative generation and OpenAI GPT-4o Mini for critical selection, ensuring high-quality output. 🗣️ Bilingual Content: Creates posts that blend English and Hindi to enhance cultural connection and broaden audience reach. 🎯 Dynamic and Contextual: Posts are automatically tailored to the specific festival of the day. 🗓️ Centralized Content Calendar: Uses Google Sheets as a reliable, easy-to-manage database for your yearly social media plan. Prerequisites 🛠️ Before you can use this workflow, you will need to: Have an n8n instance set up. Create a new, empty Google Sheet. Obtain credentials for the following services: Jina AI: Get a free Bearer Token from the Jina AI API page. Google: Set up Google credentials (OAuth2) for the Google Sheets and Google Gemini nodes. OpenAI: Get an API key from your OpenAI Platform dashboard. X (Twitter): Set up X credentials (OAuth2) to allow n8n to post on your behalf. How to use this template 🚀 🔑 Set up Credentials: In n8n, go to the "Credentials" section and add new credentials for Jina AI, Google (for both Sheets and Gemini), OpenAI, and X using the API keys and tokens you obtained. 📊 Configure the Google Sheet: Create a new Google Sheet. In the first row, create the following headers exactly as written: Name of the Festival Date English Description (Note the trailing space) Hindi Description Open the "Add all Rows at once" and "Fetch Data of Matched Date" nodes in the workflow and connect them to your Google account and select the Sheet you just created. ▶️ Populate the Data (Manual Step): Click the "Execute workflow" button on the When clicking ‘Execute workflow’ node. This will run the first part of the workflow, filling your Google Sheet with festival data for 2025. This only needs to be done once. ✅ Activate the Workflow: Save the workflow and then activate it using the toggle at the top right of the n8n canvas. The workflow will now run automatically every day to post about the day's festival. Extending the Workflow 💡 🖼️ Add Image Generation: Integrate a node like DALL-E or Midjourney to generate a unique image for each festival and include it in the tweet. 🌐 Cross-Platform Posting: Duplicate the final "Post to X" node and adapt it to post on other platforms like Facebook, LinkedIn, or Telegram. 🎨 Change the Tone: Modify the prompts in the "Generate Posts" and "Select Best Post" nodes to change the style of your social media content—make it more formal, humorous, or poetic. 📅 Use a Different Year: Update the URL in the "Get Festival Data" node to fetch data for a different year. The current URL is https://r.jina.ai/https://www.calendarlabs.com/2025-hindu-calendar.
By Harsh Maniya