RAG-powered AI voice customer support agent (Supabase + Gemini + ElevenLabs)
Ready for 3.0Ready for 3.0. This template uses no nodes that n8n 3.0 removes. No removed nodes, versions or helpers found. Static check, last run 2026-10-19.
What this workflow does
This AI workflow has 12 nodes and runs when an HTTP GET request hits its webhook URL (plus 1 other trigger). Along the way it connects Google Docs, Supabase and Google Vertex Chat Model.
How it runs
Trigger: When an HTTP GET request hits its webhook URL (plus 1 other trigger).
- Embend User Message (HTTP Request): calls an HTTP API
- Content for the Training (Google Docs): uses Google Docs
- Search Embeddings (HTTP Request): calls an HTTP API
- Splitting into Chunks (Code): runs custom JavaScript code
- Aggregate: combines items into one
- Embedding Uploaded document (HTTP Request): calls an HTTP API
- Basic LLM Chain: sends a prompt to an LLM, with model: Google Vertex Chat Model
- Save the embedding in DB (Supabase): uses Supabase
- Respond to Webhook: sends the response back to the webhook caller
Services it uses
Before you upgrade to n8n 3.0
Nothing in this workflow is removed in n8n 3.0, so it should import and run after the upgrade.
Nodes used (10)
- Aggregate
n8n-nodes-base.aggregate - Basic LLM Chain
@n8n/n8n-nodes-langchain.chainLlm - HTTP Request
n8n-nodes-base.httpRequest - Webhook
n8n-nodes-base.webhook - Respond to Webhook
n8n-nodes-base.respondToWebhook - Google Docs
n8n-nodes-base.googleDocs - Code
n8n-nodes-base.code - Supabase
n8n-nodes-base.supabase - Manual Trigger
n8n-nodes-base.manualTrigger - Google Vertex Chat Model
@n8n/n8n-nodes-langchain.lmChatGoogleVertex
From the author
Execution video: Youtube Link I built an AI voice-triggered RAG assistant where ElevenLabs’ conversational model acts as the front end and n8n handles the brain....here’s the real breakdown of what’s happening in that workflow: 1. Webhook (/inf) Gets hit by…