Automated Document Embedding and Chat Response System

This n8n workflow automates the process of transforming a Google Document into embeddings and utilizing those in an interactive chat environment. It begins with fetching document content from Google Docs, which is then split into manageable chunks. Each chunk is sent to an embedding API to generate vector representations, which are stored in a database (Supabase). The workflow also sets up a chat trigger (currently disabled) that, upon receiving a user message, generates an embedding for the message, searches for similar document chunks in the database, and combines these to provide a context-aware response. The core response generation uses a large language model (LLM) from OpenRouter, which responds based only on the retrieved document chunks, making this system ideal for knowledge-based chatbots, training assistants, or FAQ bots.