+5I want to be able to manage what text is displayed during the “Thinking” step of my AI Assistant.
Currently, the name of all the plugins being used are showing up on the screen. In some cases, these plugin names can be shown on the screen as 5 or 6 different lines, which can be noisy and look very robotic.
We’d like to have the ability to update the language used while the user waits for a response, or at the very least limit the visibility to 1 or 2 lines on the screen for all plugins.
+8I agree that this would be very helpful. Specifically, we are working right now on a PTO request plugin and have a similar experience above that
+3Hi
Fair feedback.
Currently this isn't customizable. The thinking indicator is there so users know the assistant is actively working on their request, not frozen. Without some visual feedback, it can look like nothing's happening.
That said, the direction we're heading is smarter than just swapping animations. We're working toward having the LLM generate contextual status messages, such as "Checking your timecard status...", instead of generic dots. That actually tells the user what's happening, which is more useful than what we currently do.
This is coming as the core experience continues to mature.
+2+2Already have an account? Login
No account yet? Create an account
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.