Skip to content
NEWMarketplace and your team: sell what you build, and get plans and changes that are tested before you see them. See what is new

Library / RAG & Knowledge Bases

RAG & Knowledge Bases

Sync Notion pages and Supabase embeddings for scheduled RAG retrieval

Runs on a schedule to chunk and embed Notion content into Supabase using OpenAI embeddings, keeping the retrieval store current as source data changes.

What it does

## Adjust chunk size and overlap For more accurate search results, increase the overlap. For the *text-embedding-ada-002* model the chunk size plus overlap must not exceed 8191

The steps

  1. Embeddings OpenAI
  2. Token Splitter
  3. Loop Over Items
  4. Question and Answer Chain
  5. Vector Store Retriever
  6. OpenAI Chat Model
  7. When chat message received
  8. Schedule Trigger
  9. Limit
  10. Limit1
  11. Delete old embeddings if exist
  12. Get page blocks
  13. Default Data Loader
  14. Input Reference

20 steps in total.

Similar automations