Group of 2 Use it now Notable
claude plugin test now includes a scriptable UI harness for mounting and interacting with plugin Client components
What
- A new in-process terminal UI test harness lets tests mount plugin 'Client' components and drive them with key, pointer, post, and advance actions, then unmount them, using a virtual clock, without needing a real terminal.
- The plugin-testing engine used by
claude plugin testnow exposes this as aclientsinterface withrecord,mount,act(covering drawn/key/pointer/post/advance/unmount), andreleaseAllmethods. The engine's genericcall()now feeds"ui.render"operations into this recorder, andclose()releases all mounted clients.
Why
This lets plugin authors write automated tests that render and interact with their plugin's terminal UI components directly, instead of only testing non-UI logic.
Names in the bundleui.mountact
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 Environment variables touched a line naming act after this was published.
| `CLAUDE_CODE_AUTO_MODE_SERVER` | Controls whether Claude Code asks the server to [review auto mode actions](/docs/en/permission-modes#server-side-classifier-review). When unset, Claude Code asks the server on Amazon Bedrock, Google Cloud…env-vars see the edit
Confirmed since
Anthropic's documentation has since written up act, on How Claude remembers your project.
* **You have none of those**: Claude reads every `AGENTS.md` and `.claude/AGENTS.md` in your working directory and the directories above it. In an interactive session you see a line such as `no CLAUDE.md found; AGENTS.md loaded: /home/you/…memory 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 act, on How Claude remembers your project.