[Dmitry Gozman (dgozman)]

Summary

• publish_manual.yml, run with a release branch: sets the next patch version, creates a draft release with the fixed issues (and browser versions when browsers.json changed), commits the version and publishes the packages to npm under latest-pending.
• When the draft release is published, the same workflow moves latest to the new version, after utils/check_npm_packages.sh <version> confirms that all packages are live.
• publish_release.yml does not publish to npm for a release that publish_manual.yml drafted.

Before the first use: add publish_manual.yml as an npm trusted publisher ("npm publish" and "npm dist-tag") for the 9 public packages, and cherry-pick this change with #43933, #43979 and #44009 into the release branch.

[github-actions]

Test results for "MCP"

3 failed
❌ [chrome] › mcp/crash.spec.ts:31 › crash recovery › reports the navigation error, then resets and logs the crash @mcp-ubuntu-latest-chrome
❌ [chrome] › mcp/crash.spec.ts:58 › crash recovery › lists only one tab @mcp-ubuntu-latest-chrome
❌ [chrome] › mcp/crash.spec.ts:79 › crash recovery › marks non-current crashed tab in the tab list @mcp-ubuntu-latest-chrome

5232 passed, 631 skipped

---

Merge workflow run.

[github-actions]

Test results for "tests 1"

23628 passed, 713 skipped

---

Merge workflow run.

[microsoft-playwright-automation]

🟡 Crash-recovery failures are uncertain; the runner changed

Hi, I'm the Playwright bot. I checked the failures; the release-only diff does not reach these tests.

Details and evidence

This covers MCP run 43249898116, attempt 1, head d2621691c80fe106898ae319801edcfa28b36422. The merged report lists three failures, all on Ubuntu Chrome.

Uncertain

• mcp/crash.spec.ts:31, reports the navigation error, then resets and logs the crash — reaches the navigation-error check, then times out waiting for the reset and crash log.
• mcp/crash.spec.ts:58, lists only one tab — times out after chrome://crash, then reports Not connected.
• mcp/crash.spec.ts:79, marks non-current crashed tab in the tab list — the same timeout and Not connected error for the other tab.

The failed job used Ubuntu 26.04.1. The earlier main job, at bac1f9f67e9cff2a8e7ecbac25d56ec3bb1d15c6, used Ubuntu 24.04.5. The existing skip in crash.spec.ts:22 excludes only ubuntu24.04, with the reason “never dispatches the crash event”; it does not exclude 26.04.

The CI history snapshot, updated with this job's results, shows each test skipped in 726 earlier Ubuntu Chrome runs, with no earlier failures on any recorded browser/platform. Each passed in 723 macOS Chrome runs and 726 Windows Chrome runs. That proves neither a flake nor correct behavior on Ubuntu 26.04.

The runner change is the likely explanation, not this PR: MCP code, helpers, browser versions and its CI workflow are unchanged. Confirming the cause requires these tests on the base commit under Ubuntu 26.04. If that confirms the existing crash-event limitation, extend the skip to ubuntu26.04; do not change the release workflow.

_{Checked the failures on Linux by the Playwright bot - agent run}