Follow Discord
Sweep 10 Nov 2026 · 02:18Z Build v2.1.322 539 read Stable v2.1.311 Latest v2.1.322 Next v2.1.322 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.310 ·

Claude Test gains a progress panel in the terminal

The built-in Claude Test plugin gets a terminal panel showing test progress, a pane note tool, and new settings for its browser tooling

Group of 3 You'll notice Internal Changes
JSON All of v2.1.310
You'll noticeTier: how much it should matter to you
5Useful: my rating, 1 to 5
4Signal: worth watching, 1 to 5
MCPArea: what it touches
Internal ChangesKind: in v2.1.310,
What probably matters to youSection of the release

What

Claude Test is a bundled plugin for testing an app in a browser. It gains a terminal panel and supporting pieces:

  • A new "Claude Test" panel shows app start and ready states ("Starting your app…", "Opening your app…"), "Read your code", "Writing the tests…", a row per test with pass, fail or blocked marks, a progress bar, an estimate of time left, and hints such as "/claude-test fix" and "Details in the chat · ". An "Open the full report" label was also added. The command name /claude-test appears in this panel code and did not appear before.
  • Claude is told when the panel is open in the terminal and that the skill's panel rule applies, with a variant for terminals that cannot draw screenshots.
  • The build also contains a built-in plugin id claude-test@builtin, agents claude-test:runner and claude-test:explorer, run folders under .claude-test/runs, and an entry pairing claude-test with cc-plugin-claude-test.
  • A new tool, mcp__plugin_claude-test_browser__claude_test_pane_note, was added to two allow lists.
  • The built-in Playwright MCP server (the browser-control add-on) now gets CLAUDE_TEST_ESSENTIAL_TRAFFIC=1 when a condition holds. Before, it got only CLAUDE_TEST_PROJECT_DIR.
  • A "ui.blit" entry appears in a method allow list.

No gate controlling the panel was traced.

Why

Claude Test is gaining a terminal view of a test run, so progress and results can be followed without reading the chat.

See this entry in the whole of v2.1.310 →

Feedback