I checked everywhere and didn’t find answers.
Describe the problem/error/question
I don’t know how to realize my idea, have you got an idea or an exemple ?
I’d like to have a chatbot that:
- Access 2 PostgreSQL tables
- Can modify one of them (Add or Modify)
- Scalable to 1.6k persons (not at same time)
- The business application is: shop managers are not giving informations to HQ about their future plans, it would be too time consuming. Everything is in their head. Thus giving them a chatbot to talk will make things possible. They talk => chatbot update database about future business state.
I am connected to chatGPT4o and Supabase (for Postgre)
I have tried:
- AI agent SQL but I can’t figure out how it works nor a source to show me
- AI agent with 3 Postgre tools: consultDB1, consultDB2, modifyDB2
- It struggles to modify correctly DB2 thus in don’t want to add a 4’th tool to add records - AI agent with 3 Postgre modules: consultDB1, consultDB2, modifyDB2
- I didn’t succeed to get standard results out of the agent to trigger modules and then come back to chat
I am beginner, followed lessons and doing this as a proof of value only, thus no need for voice API now, just update the database.
What would you do ? have you got examples ?
Please share your workflow
I have tried/deleted to much and didn’t save it.
Information on your n8n setup
- n8n version: 1.67.1.
- Database (default: SQLite): PostgreSQL
- n8n EXECUTIONS_PROCESS setting (default: own, main): Pas trouvé donc je dirais “own”
- Running n8n via: cloud
- Operating system: windows10 home