How to get raw system prompt that actually gets sent to the LLM

[tannisroot]

Maybe I'm missing something, but is there a way to get a raw system prompt message that gets sent to the LLM when you use the Assist pipeline, with all the devices and services added to it and not just "{{ services }}" and "{{ devices }}?
The reason I'm asking is I want to look at how different parameters and changes in the prompt affect the model's response, and doing so through the assist pipeline is not very convenient (plus it hides some of the output), so I figured I would just simulate it with a modelfile.
I tried looking at Ollama's logs hoping it prints it there, but unfortunately it does not :(

[acon96]

If you turn on debug logging for the component in your Home Assistant's /config/configuration.yml it will print the prompt to the HA log when you're chatting.