Overriding instructions with Function Calling?

Hey guys. I feel like I’m missing something here. When GPT(Assistants) decides to call a function I am required to respond using https://api.openai.com/v1/threads/{thread_id}/runs/{run_id}/submit_tool_outputs

But I want to also override the instructions, an option given when creating a new run object:

string or null
Override the default system message of the assistant. This is useful for modifying the behavior on a per-run basis.

For my use-case I want GPT to respond with only a formatted string (for myself to update a field with), not the typical verbiage that it spews before & after. The function is calling for the user profile and then I want GPT to perform some work on the information and then return simply saying “Done! How does that look?”

What am I missing here? Surely there’s a way to set new instructions once GPT decides a function needs to be called? The only option I can see is to use the old endpoints and call a separate Completion GPT, but… it just feels wrong.

EDIT: I just decided to call a separate GPT and return an “OK”. Thinking further yeah, it makes sense. lol