A department-aware support bot that answers from your own documents

A messenger bot answers customer questions from your own documents — separately for billing, returns, and technical issues — and the knowledge updates simply by dropping a file into a folder.

  • Answers come from your documents, not from general knowledge
  • Each department has its own document set, so answers don't blur together
  • Updating the knowledge means dropping a file in a folder — no reconfiguring
  • Conversation history builds up in a database, so you see what's asked most
Pinecone Vector StoreGoogle DriveEmbeddings CohereDefault Data LoaderPostgres
Connect Qdrant Vector Store and launchScriptera

How it works

To keep Slack and Qdrant Vector Store in sync, connect them with a ready-made scenario — changes flow automatically.

  1. Starts when: Google Drive Trigger
  2. Then: Download file
  3. Then: Pinecone Vector Store3
  4. Check: Switch
  5. If yes: return policy
  6. Then: return policy1
  7. If no: talk technical
  8. Then: tech questions
  9. If no: billing
  10. Then: billing1
  11. If no: Send a text message3
  12. Then: Embeddings Cohere3
  13. Then: Default Data Loader
  14. Then: Character Text Splitter
  15. Then: Pinecone Vector Store4
  16. Starts when: Google Drive Trigger1
  17. Then: Download file1
  18. Then: Embeddings Cohere4
  19. Then: Default Data Loader1
  20. Then: Character Text Splitter1
  21. Then: Pinecone Vector Store5
  22. Starts when: Google Drive Trigger2
  23. Then: Download file2
  24. Then: Embeddings Cohere5
  25. Then: Default Data Loader2
  26. Then: Character Text Splitter2
  27. Starts when: Telegram Trigger
  28. Then: Execute a SQL query1
  29. Then: Select rows from a table
  30. Check: Switch1
  31. If yes: Send a text message
  32. Then: Execute a SQL query3
  33. If no: Send a text message4
  34. If no: AI Agent3
  35. Then: Send a text message1
  36. If no: Execute a SQL query2
  37. Then: Send a text message2
  38. Then: Pinecone Vector Store6
  39. Then: Embeddings Cohere6
  40. Then: OpenRouter Chat Model3
  41. Then: Simple Memory3
  42. Then: Pinecone Vector Store7
  43. Then: Embeddings Cohere7
  44. Then: Pinecone Vector Store8
  45. Then: Embeddings Cohere8
  46. Then: Execute a SQL query

You can launch this Slack + Qdrant Vector Store integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios