A knowledge base built from your document folder: ask, get an answer

The scenario turns the documents in Microsoft OneDrive into a searchable memory in Pinecone, keeps a registry in таблица, and then answers questions from your own documents — with the source attached.

  • Answers come from your documents, not from imagination
  • A registry in таблица shows exactly what's been processed
  • The same documents never get processed twice
  • An email to AWS SES tells you when ingestion is done
Google SheetsPinecone Vector StoreDefault Data LoaderEmbeddings OpenAI
Connect Pinecone and launchScriptera

How it works

To move data from AWS SES to Microsoft OneDrive automatically, use a ready-made scenario — no manual exports.

  1. Starts when: Manual Ingestion Trigger
  2. Then: List Knowledge Base Docs
  3. Then: Download Doc as Text
  4. Then: Store Embeddings in Pinecone
  5. Then: Log to Document Registry
  6. Then: Notify Ingestion Complete
  7. Then: Read Document Registry
  8. Then: Detect New or Updated Docs
  9. Check: IF New Docs Found
  10. If yes: Prepare Docs for Re-ingestion
  11. Then: Download Updated Docs
  12. Then: Re-ingest into Pinecone
  13. Then: Update Document Registry
  14. Then: Build KB Digest Email
  15. Then: Send Weekly KB Digest
  16. If no: Build KB Digest Email1
  17. Then: Send Weekly KB Digest1
  18. Then: Load Document Data
  19. Then: Split Into Chunks
  20. Then: Generate Embeddings
  21. Starts when: Question Webhook
  22. Then: Extract Question & User
  23. Then: Search Knowledge Base
  24. Then: Build RAG Prompt
  25. Then: Answer Generator
  26. Then: Parse AI Answer
  27. Then: Log Q&A to Sheets
  28. Starts when: Send Answer Response
  29. Then: Generate Query Embedding
  30. Starts when: Weekly Sunday 11AM Trigger
  31. Then: List Knowledge Base Docs1
  32. Then: Load Document Data1
  33. Then: Split Into Chunks1
  34. Then: Generate Embeddings1

You can launch this AWS SES + Microsoft OneDrive integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios