What
Plugins can ask Claude for an answer through a function called model.complete. That function can now take an optional abort signal, which is a way to tell a call to stop. If the signal fires before or during the call, the call does not throw an error. It returns a result that says the question was not answered, gives the reason as "aborted", and reports empty usage.
Why
Plugin authors can stop a model call part way through and handle that like any other outcome, without catching an error.
Names in the bundlemodel.complete
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
Confirmed since
Anthropic's documentation has since written up model.complete, on Claude Code changelog.
* Added prompt caching to `$.model.complete` for mods: `prompt` and `system` take blocks of text, and `cache: true` on a block caches the request up to itchangelog see the edit
One source agreesOne thing we can check says the same as this entry.
Anthropic's documentation agrees
Anthropic's documentation has since written up model.complete, on Claude Code changelog.