Hi guys, I’m a developer and new to n8n with supabase vector database. I have a project that user can chat with AI Agent, then vectorize the extracted data, then go into the supabase vector database to find which data has the highest similarity score and return the result to the user. But currently, the agent does not check the data in the vector database but only uses its existing knowledge to respond to the user, please check if there is any problem in my flow
here is my workflow
I’ve made a few adjustments to the prompts.
Can you try again with these settings, please?
See if the AI is activating the Vector Store node.
Thanks for your reply. It activating the Vector Store node but it not return any data for me, I guess my match_documents query name did not works well
Oh, I see. So the setup inside Supabase might be wrong.
Take a look at this Supabase documentation.
Also, check out this video from Cole Medin:
If my reply answers your question, please remember to mark it as a solution.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.