What
Child agents spawned by claude plugin eval now run detached and are killed as a whole process group on completion, interrupt or timeout, so tools they spawned do not outlive the run. The runner also accepts a separate set of operator-forced allowed tools, merged with the ones the eval case declares.
Details
- Non-Windows kills the group with SIGKILL by negative process id; Windows uses
taskkill /T /F. - Detaching is skipped on Windows.
Evidence
process.kill(-Ce.pid, "SIGKILL")
Strings lifted out of the shipped bundle, so the claim above can be checked against them.
Names in the bundleclaude plugin eval
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.
Added since
A small documentation edit on Create and distribute a plugin marketplace touched a line naming claude plugin eval after this was published.
Test your marketplace before sharing. Validation checks file structure; to test whether a plugin changes what Claude does on realistic prompts, run its eval suite with [`claude plugin eval`](/docs/en/plugin-evals) before you publish a new …plugin-marketplaces see the edit
Confirmed since
Anthropic's documentation has since written up claude plugin eval, on Claude Code changelog.
* Added `claude plugin eval`: run a plugin's eval suite against Claude Code and get scored, reproducible results (JSON + HTML report); see `claude plugin eval --help`changelog 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 claude plugin eval, on Claude Code changelog.