Messenger support: answers from your knowledge base, calls a human in time

The scenario reads every message for intent and mood, answers from your knowledge base and orders sheet, and hands complaints or requests for a human straight to a person.

  • The customer's mood is detected — an escalation goes to a human at once
  • Answers come from your knowledge base and orders, nothing invented
  • Voice notes are understood just like text
  • The conversation's context is kept — next time it continues rather than restarts
Postgres Chat MemoryGoogle Docs ToolGoogle Sheets ToolSupabase ToolWhatsApp
Set it upScriptera

How it works

To move data from Google Cloud Firestore to Google Sheets automatically, use a ready-made scenario — no manual exports.

  1. Starts when: WhatsApp Trigger
  2. Check: Voice or Text
  3. If yes: Download media
  4. Then: Transcribe1
  5. Check: Merge
  6. Then: intent classifier
  7. Check: Merge4
  8. Then: Aggregate
  9. Then: Human intervention
  10. Then: Owner escalation
  11. Then: Human request.
  12. Then: Critical complaint.
  13. Check: Merge3
  14. Then: Aggregate3
  15. Then: Customer service AGENT
  16. Then: Historial Chat and feedback
  17. Then: Aliment context for next messages
  18. Then: Send message
  19. Then: sentiment classifier
  20. Then: privacy classifier
  21. Then: Generate conv context
  22. If no: Text
  23. Then: OpenAI Chat Model8
  24. Then: OpenAI Chat Model10
  25. Then: OpenAI Chat Model11
  26. Then: Think1
  27. Then: Structured Output Parser
  28. Then: Structured Output Parser1
  29. Then: Postgres Chat Memory
  30. Then: Structured Output Parser2
  31. Then: Knowledge base
  32. Then: orders database
  33. Then: OpenAI Chat Model
  34. Then: OpenAI Chat Model1
  35. Then: OpenAI Chat Model2
  36. Then: Supabase1

You can launch this Google Cloud Firestore + Google Sheets integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios