Supabase Retrieve Documents as Vector Store. Metadata filter not working?
I am working on a Slack chatbot. I am using Supabase to store the chats and the documents. I added a field to the metadata called: “thread_ts” this adds the corresponding time stamp. So you can have different threads in Slack working with the AI Agent node. When the “Slack Trigger” node is triggered a thread_ts value is sent with it. Now I only want the corresponding documents and chats be retrieved by the AI Agent.
I use this flow: AI Agent - Tool - Answer and Questions with Vector store - Supabase Vector Store.
In the “Supabase Vector Store” node there is a “Metadata Filter” I add a field name and then i need to enter an expression to get the corresponding thread_ts. However there is no input in this node. So how would I be able to put that value in there? I tried a few basic expressions like: “{{$json.thread_ts ? $json.thread_ts : $json.ts}}” but it gives me an empty array as a result.
Please share your workflow
Share the output returned by the last node
no output.
Information on your n8n setup
- n8n version: 1.114.0
- Database: Supabase
- n8n EXECUTIONS_PROCESS setting default:
- Running n8n via n8n cloud
- Operating system: Windows 11
