Hi all!
I have been testing an assistant and I noticed in the debug and in the run history that sometimes it defaults to “completed” every time it responds. Other times it shows “awaiting_response” like it is supposed to.
The “completed” status is an issue because it slows down the response rate and triggers a flow too early.
I have rephrased the instructions and the goal for the assistant a few times but notice it keeps happening. Any tips on why this might be happening?
The image below shows two chats that are active at the time. (The users are on zoom with me so I can see their screen). One is complete (even though it is still running, albeit slowly).
Thanks all!