Automated Storage of Notion Pages as Vectors in Supabase

This workflow automates the process of capturing Notion pages and storing them as vectorized documents in a Supabase database, enabling efficient retrieval and AI-driven analysis. It begins with a trigger on new pages added to Notion, then retrieves and processes page content by filtering out media files, concatenating text blocks, and generating embeddings via OpenAI. The text is split into manageable chunks for better embedding quality, and metadata such as page IDs and timestamps is created for context. Finally, the processed text and embeddings are stored in a Supabase table with a vector column, making this workflow ideal for building semantic search or AI-powered knowledge bases based on Notion content.