A clinic assistant in your messenger: answers, bookings, hand-off to a human

The assistant answers patients from your own material, books free slots, reschedules and calls in a receptionist when a human is needed. Everything is collected into a sheet.

  • Answers from your own material on services and prices
  • Understands text, voice notes and photos
  • Books and reschedules appointments
  • Calls in a receptionist when a human is needed
Data table ToolPinecone Vector StoreDefault Data LoaderGoogle DocsEmbeddings OpenAI
Connect Postgres PGVector Store and launchScriptera

How it works

To move data from Microsoft SQL to Postgres PGVector Store automatically, use a ready-made scenario — no manual exports.

  1. Starts when: Schedule Trigger
  2. Then: Get slots
  3. Then: Append or update slot
  4. Then: Get lead
  5. Then: Append or update lead
  6. Then: Get human call
  7. Then: Append or update human call
  8. Then: Get human appointment
  9. Then: Append or update appointment
  10. Then: Booked Appointment
  11. Then: AI Summarizer1
  12. Then: Format output
  13. Then: Insert Lead
  14. Check: Check Appointment1
  15. If yes: save appointment - same service
  16. Check: If Appointment is ready
  17. If yes: save appointment - different service
  18. Then: AI Data Extractor
  19. Then: Output Format Text
  20. Check: Check Human is called
  21. If yes: Send Email Notification If Human is called
  22. Then: Add row for Human Call
  23. Then: Send via WhatsApp
  24. Then: Pinecone Vector Store
  25. Then: Default Data Loader
  26. Then: Recursive Character Text Splitter
  27. Then: Get a document
  28. Starts when: When clicking ‘Execute workflow’
  29. Then: Embeddings OpenAI
  30. Then: Vector Store Tool
  31. Then: Pinecone Vector Store (Retrieval)
  32. Then: Embeddings OpenAI1
  33. Starts when: Verify • WA (hub.challenge)
  34. Check: filter text messages
  35. If yes: Prepare Text Data
  36. Then: Insert Row Memory
  37. Then: Get History
  38. Then: Get only users appointment
  39. Then: Format Context
  40. Then: Main AI Agent
  41. Then: Update Row
  42. Then: Get leads
  43. Check: Check if Lead
  44. Check: If Audio
  45. If yes: Get Audio
  46. Then: Download Audio
  47. Then: Transcribe Audio
  48. Check: If Image
  49. If yes: Get Image
  50. Then: Download Image
  51. Then: Analyze image
  52. Then: Edit Fields
  53. Check: If Video/PDF
  54. If yes: Get file
  55. Then: Download file
  56. Then: Prepare Email With Attachment
  57. Then: Send Email with attachment
  58. Then: Simple Memory
  59. Starts when: WhatsApp Webhook
  60. Then: OpenAI Chat Model

You can launch this Microsoft SQL + Postgres PGVector Store integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios