Fleet Command Done Spawned Too Hard!

[ChadChapman]

Describe the bug

/feedback I tried out the new fleet command, and looks like it blew out the requests rate.
I sent it out to just write some JSDocs for two small-medium size dirs.
I let Copilot handle the fleet propagation aspects, and lesson learned on that, haha.
For this feature, it would be nice to defensively restrict how many sub-agents the fleet can spawn, and aggressively throttle rather than let them "go for it", up until hitting rate limits and getting pooched for 4 hours :(.
Life of the YOLO dev living on the bleeding edge I suppose ;).
Thanks for reading this, it is appreciated!
Have a great day, great night, and a real nice week!
Thank you!

Result: quite a few of these:

● Background agent "Generating JSDoc for manifest bridge" (jsdoc-documenter) completed 
  └ Generate comprehensive JSDoc comments for all public APIs, props, and helper ...

✗ Sorry, you've hit a rate limit that restricts the number of Copilot model requests you can make within a specific time period. Please try again 
in 
  4 hours. Please review our Terms of Service (https://docs.github.com/site-policy/github-terms/github-terms-of-service). (Request ID: 
  E64C:1ABCE0:69831E:7CAFE4:69D48881) 

❯ ok, our fleet got too big, commit all changes in the worktree(s) and wrap it up

✗ Sorry, you've hit a rate limit that restricts the number of Copilot model requests you can make within a specific time period. Please try again 
in 
  4 hours. Please review our Terms of Service (https://docs.github.com/site-policy/github-terms/github-terms-of-service). (Request ID: 
  E838:245D28:5A0769:69F6F8:69D488F0) 

Background tasks basically all did the rate-limit thing, in tandem.
Never hit a rate limit with Copilot CLI, and I been here since the OG legacy version, pre-current-fancy cli.

Affected version

GitHub Copilot CLI 1.0.19.

Steps to reproduce the behavior

Entire, complete, and only command I done ran that got us here:

/fleet  what to do:
  1. create a new worktree, named `@copilot/cli/gpt-4.1/jsdoc-documenter/document-react-components-1`
  2. in a background terminal/process, etc checkout and use the newly generated worktree
  3. audit all files under @site/src/components/ and under @site/src/pages/ for JSDoc resources/entries
   a. at both the file-level, and for all helpers, and all components, also for any and all other resources/file contents where JSDoc
resources
  enhance/extend/improve DX, readability, clarity, etc.
  4. only use valid/recognized/official JSDoc fields/attributes/tags.
   a. refer to official docs site at `https://jsdoc.app/`, or context7 resources as needed
   b. do not invent or freestyle new/unofficial/unrecognized JSDoc tags.  these will fail validation and are considered anti-patterns and will
  require further work after they fail validation checks then have to be redone/fixed.
  5. use fleet for multi agent orchestration(s), for parallel executions, as these should not require any serialization of steps, etc.
6. in fleet mode, if it is preferred/SOP/expected approach for each subagent to then use a dedicated sub-worktree, off the initial worktree,
that is fine, just keep me informed.

Expected behavior

First time using /fleet.
Not sure exactly what to expect.
Do expect at least a warning about fleetin' too hard, and a heads-up on getting rate limited.

Additional context

No response

[seat-01]

Maybe we should push for "Rate Throttle" as kind of a workaround:

• Replace "Rate Limit" with "Rate Throttle" #2738