An assistant that labels your inbox and learns from your answers

The scenario reads every new email and applies a label. When it isn't sure, it asks you in chat and remembers your answer as a rule. Over time it asks less.

  • Labels are set by what the mail means, not by rigid filters
  • Borderline mail is sent to you for confirmation
  • Every decision you make becomes a rule for next time
  • Rules can be deleted — the assistant stops applying them
Data table ToolGmail
Set it up for yourselfScriptera

How it works

To connect devino and Data table Tool, you don't need a developer: a ready-made scenario links them in minutes.

  1. Starts when: When Executed by Another Workflow
  2. Then: Gmail backfill
  3. Then: Get labels
  4. Check: filter for AI labels
  5. Then: Add label map
  6. Then: set confidence threshold
  7. Then: AI Check Email
  8. Then: add label & email metadata to ai output
  9. Check: If low confidence...
  10. If yes: Add rule for review
  11. Then: Add AI recommended labels
  12. Then: Add fixed labels
  13. Then: Telegram: Error
  14. Then: Unwanted Email Output Parser
  15. Then: OpenAI Chat Model
  16. Then: OpenAI Chat Model1
  17. Then: Previous Rules and actions tool
  18. Starts when: When clicking ‘Execute workflow’
  19. Then: Get all labels
  20. Check: get AI labels
  21. Then: Get all pending rules
  22. Check: Loop Over Items
  23. Then: Telegram: Train
  24. Check: Switch
  25. If yes: Update rule
  26. Then: Message: Added rule
  27. Then: Remove label: AI, Needs Review
  28. If no: Delete rule
  29. Then: Message: Deleted rule
  30. Starts when: Gmail Trigger

You can launch this devino + Data table Tool integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios