{"version":"2.1.302","anchor":"claude-test-plugin-bundle-with-explorer-author-and-runner-a","canonical_anchor":"claude-test-plugin-bundle-with-explorer-author-and-runner-a","heading":"Bundled Claude Test plugin runs browser-based tests with three helper agents","tier":"use","area":"Plugins","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.302\/e\/claude-test-plugin-bundle-with-explorer-author-and-runner-a","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.302","markdown":"### Bundled Claude Test plugin runs browser-based tests with three helper agents\n\nClaude Code bundles a Claude Test plugin, run with `claude-test`, with author, explorer and runner agents that drive a browser\n\n**Unclear.** What changed in the plugin, and whether it is available to users, is not stated.\n\n**What**\n\nClaude Code bundles a plugin called Claude Test. It is started with the `claude-test` command and its skills, and it ships three agents, which are helper copies of Claude with set jobs:\n\n- author\n\n- explorer\n\n- runner, which works in the background, drives the app being tested in the plugin's own closed-off browser with no visible window, and judges test descriptions\n\nIt controls the browser through tools named `mcp__plugin_claude-test_browser__*`. The plugin was in the previous build and has changed in this one.\n\n**Why**\n\nIt brings a browser-based testing workflow into Claude Code.\n\n- Area: Plugins\n- Names: `claude-test`\n- Tier: Use it now\n- Useful: 4\/5\n- Signal: 3\/5\n- Scope: individual\n- Heads-up: no"}