Follow Discord
Sweep 22 Oct 2026 · 04:32Z Build v2.1.308 527 read Stable v2.1.308 Latest v2.1.308 Next v2.1.308 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.305 ·

Claude Test: a safer browser helper and a new all argument

The /claude-test skill's browser helper ignores risky project environment settings and explains missing Node.js, and the skill accepts all

Group of 2 Use it now Improvements
JSON All of v2.1.305
Use it nowTier: how much it should matter to you
2Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
SkillsArea: what it touches
ImprovementsKind: in v2.1.305,
What probably matters to youSection of the release

Unclear Whether this test skill is available to users is not known.

What

Claude Test is a built-in skill run with /claude-test. A skill is a packaged set of instructions Claude can follow. Two parts of it changed.

  • The browser helper is a Playwright MCP server (playwright-mcp.mjs), a small program that lets Claude drive a web browser. It now starts with a cleaned-up set of environment variables (settings passed to programs when they start). Values for proxy variables, NODE_OPTIONS, LD_PRELOAD and similar ones are not taken from the project's .claude/settings.json or .claude/settings.local.json, and a warning says so.
  • The helper now checks the path to Node.js before using it. Setup results are recorded as diskless, no_node, linked, differs or unwritable.
  • A new no_node error appears when no working Node.js is found or Node.js takes too long to answer. It tells you to install Node.js or set a default Node version in asdf or mise, then restart and run /claude-test.
  • The skill's argument hint now reads [app folder] [all | words from a spec name, to run only those | fix | onboard], adding all. The skill's bundled instructions file also changed.

Why

A project's settings files can no longer slip values such as NODE_OPTIONS or LD_PRELOAD into the browser helper, and a missing Node.js now produces a clear instruction instead of a silent failure. The new all argument appears to run every spec, though the members only show the hint text.

How sure we are
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubtWhether this test skill is available to users is not known.

See this entry in the whole of v2.1.305 →

Feedback