Generate Daily Content Ideas with Webhook, Supabase, and RAG Agent

Overview

This n8n workflow automates the generation of daily content ideas by leveraging a webhook trigger and a sophisticated RAG (Retrieval Augmented Generation) agent. The process begins when the Webhook Trigger receives a request, initiating a chain of actions. The incoming data is first processed by the Text Splitter to break it into manageable chunks, followed by the Embeddings node to create vector representations of this text. These embeddings are then inserted into a Supabase vector store using Supabase Insert. Simultaneously, a Supabase Query node retrieves relevant context, which is then utilized by the Vector Tool to inform the RAG Agent. The RAG Agent, powered by an OpenAI Chat Model and Window Memory, generates new content ideas based on the retrieved information and conversational history. Finally, the generated ideas are logged to a Google Sheet via the Append Sheet node, with potential errors reported to Slack by the Slack Alert node.