Handle WhatsApp support chats with OpenRouter, Pinecone, and Gemini

by Salman Mehboob ✓ · 7,154 views · created 2026-07-13 · Support Chatbot, AI Chatbot

What this workflow does

This AI workflow has 44 nodes and runs on WhatsApp events (messages) (plus 2 other triggers). Along the way it connects WhatsApp Business Cloud, OpenRouter Chat Model, HTTP Request and Pinecone Vector Store and 1 more. It also calls host.docker.internal:3001 over HTTP.

How it runs

Trigger: On WhatsApp events (messages) (plus 2 other triggers).

1. Route by Event (Switch): routes items into 3 branches
2. Fetch Messages (HTTP Request): calls an HTTP API
3. Has Media? (If): checks a condition and branches into true / false
4. Store Read (HTTP Request): calls host.docker.internal:3001 (POST)
5. Store Sent (HTTP Request): calls host.docker.internal:3001 (POST)
6. Store Delivered (HTTP Request): calls host.docker.internal:3001 (POST)
7. Contact Lookup (HTTP Request): calls an HTTP API
8. Summarize (Basic LLM Chain): sends a prompt to an LLM, with model: OpenRouter Model
9. Route By Media Type (Switch): routes items into 4 branches
10. Send Text (WhatsApp Business Cloud): send

…plus 25 more steps. See the full node list below.

Services it uses

WhatsApp Business CloudOpenRouter Chat ModelHTTP RequestPinecone Vector StoreEmbeddings Google Gemini

HTTP calls to: host.docker.internal:3001

Before you upgrade to n8n 3.0

Nodes used (15)

• WhatsApp Triggern8n-nodes-base.whatsAppTrigger
• Switchn8n-nodes-base.switch
• HTTP Requestn8n-nodes-base.httpRequest
• Edit Fields (Set)n8n-nodes-base.set
• Ifn8n-nodes-base.if
• WhatsApp Business Cloudn8n-nodes-base.whatsApp
• OpenRouter Chat Model@n8n/n8n-nodes-langchain.lmChatOpenRouter
• AI Agent@n8n/n8n-nodes-langchain.agent
• Simple Memory@n8n/n8n-nodes-langchain.memoryBufferWindow
• Http Request Tooln8n-nodes-base.httpRequestTool
• Pinecone Vector Store@n8n/n8n-nodes-langchain.vectorStorePinecone
• Embeddings Google Gemini@n8n/n8n-nodes-langchain.embeddingsGoogleGemini
• Webhookn8n-nodes-base.webhook
• Respond to Webhookn8n-nodes-base.respondToWebhook
• Basic LLM Chain@n8n/n8n-nodes-langchain.chainLlm

From the author

Quick overview This template implements a WhatsApp support suite that logs inbound events to a dashboard API, routes conversations through an OpenRouter-powered AI agent with Pinecone RAG and memory, exposes a webhook for human outbound replies, and provides…

View the original on n8n.io → Check your own copy