Endform logo

Skipped tests and Playwright locks

Skipped tests now show up across the dashboard, and Endform respects the new Playwright 1.63 test locks, with per-lock limits you can set in endform.config. Quarantine moves to its own dashboard tab for every organization. CLI 0.81 also supports more playwright test flags, including --trace, --fully-parallel, --timeout and --add-reporter.

Major features & improvements

  • Skipped tests in the dashboard: Skipped tests now appear in dashboard counts, filters, history and test details, and don’t count towards pass rate or duration. Tests skipped while running still run their teardown. Tests skipped during discovery are recorded without starting an attempt.
  • Playwright 1.63 test locks: Endform reads the locks on your tests and schedules them the way Playwright does, so tests that share a lock never run at the same time. Use a lock:<name> label in concurrentTestLimits to change a lock’s limit, or to apply it within one suite run or across all runs.
  • Quarantine tab: Quarantine now has its own tab in the dashboard for all organizations, replacing the old settings page. You can edit existing rules instead of recreating them. Each rule shows a recommendation based on the last 7 days of results: keep it, deactivate it once the test passes reliably, or wait for more history. Duplicate rules are flagged too. Rules can also be edited from the CLI and the MCP server.

CLI version updates

Upgrade with npx endform@latest or your usual install method. The latest version this week is 0.81.x.

Version 0.80.x

  • endform live-test: endform debug-test is now endform live-test, and the matching MCP tools and environment variables have been renamed too. Older CLIs are asked to upgrade.

Version 0.81.x

  • Skipped test reporting: The CLI reports runtime skips and tests skipped during discovery, and marks skipped attempts with ~ in the run output. With older CLIs, runtime skips show as passed in the terminal and discovery skips don’t reach the dashboard.
  • Test locks and lock limits: Playwright 1.63 test locks are picked up during suite analysis, and concurrentTestLimits accepts lock:<name> labels.
  • More Playwright flags: --trace is respected, --fully-parallel is respected, -x stops after the first failure, -G is short for --grep-invert, and --timeout <ms> overrides every project’s test timeout, up to the 240-second limit.
  • endform quarantine update: Change an existing quarantine rule’s test, project, branch or reason from the terminal.