Extracting a URL

I want to give users the option to access a pdf on a website. So I have a txt file in the knowledge base with some 70 URLs. Each URL has a unique code in it of the format 'AB123'.

I prompt the user to enter that code and then want to extract the appropriate URL so that I can send them to the document. I have tried to extract using AI response but it returns inconsistent values.

Can anyone suggest a suitable approach?

Thanks