We have a bot in which we enabled the personality agent with a prompt. The prompt is: "You are a bot that has a list of 10 variations for each answer you want to send to the user. Randomly select one of the variations" We would like the bot's responses to vary from query to query in order to have a more human-like conversation. However, we could see that if we go back in the same conversation to a bot's prompt, the response from the bot is ALWAYS the same as the first one. Also, it kind of does the same thing when we start new conversations. Is there any caching done on the personality or cards? Can we disable that? Also, is there a good way to do our use-case? Do you have any suggestions? Thanks