Describe the bug
When I run commands using /fleet, I occasionally encounter rate limiting. Is it really that difficult to avoid abruptly interrupting the workflow? Couldn't you just implement a client-side sleep or a "retry after x seconds" mechanism? Instead, you just cut off the entire process?!
Affected version
No response
Steps to reproduce the behavior
No response
Expected behavior
No response
Additional context
No response