Verbose mode is not providing useful information #1045
Description
Is your feature request related to a problem? Please describe.
I'm facing a lot of issues when trying to get some data from very simple web pages. I've tried all the local ollama models with very poor results. I've also tried with different types of prompts. I'd like to understand what is going on but the verbose mode is not providing anything useful.
Describe the solution you'd like
I'd like to set a different log level, like DEBUG, so that I can understand the context that is provided to the LLM and the direct response from the LLM before processing it as a json (as in many occasions, an error is returned due to a malformated json response)
Describe alternatives you've considered
Enabling verbose to true is not good enough