How to force RAG agent to only use the data from the VectorStore and not think on its own ? #73
-
|
The title pretty much says it all. I am building a RAG agent to answer questions. I have a data store of frequent question and i want the LLM to only build his answer from these informations. Is there a simple way to do this ? |
Beta Was this translation helpful? Give feedback.
Answered by
ilvalerione
May 9, 2025
Replies: 1 comment
-
|
The first thing I would do is an authoritative and detailed system prompt where you instruct the agent on what you want it to do, what you want it not to do, how it should act relative to the information in the context, and most importantly why. Prompt engineering typically work in this use case. |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
obaume
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The first thing I would do is an authoritative and detailed system prompt where you instruct the agent on what you want it to do, what you want it not to do, how it should act relative to the information in the context, and most importantly why. Prompt engineering typically work in this use case.