feat(webkit): roll to r2366

[microsoft-playwright-automation]

Browser build: https://github.com/microsoft/playwright-browsers/commit/cb092a970a8382d1f393b9abfa9521ec70a9e61f. Build has full platform coverage - check that all tests pass there.

[github-actions]

Test results for "tests 1"

5 flaky

⚠️ [chromium-library] › library/chromium/chromium.spec.ts:301 › should report intercepted service worker requests in HAR `@realtime-time-library-chromium-linux`
⚠️ [chromium-library] › library/chromium/chromium.spec.ts:373 › should produce network events, routing, and annotations for Service Worker `@chromium-ubuntu-22.04-node24`
⚠️ [chromium-library] › library/tracing.spec.ts:533 › should produce screencast frames fit `@chromium-ubuntu-22.04-node22`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:257 › third party 'Partitioned;' cookies `@firefox-ubuntu-22.04-node20`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:470 › top level 'Partitioned;' cookie and same origin iframe `@firefox-ubuntu-22.04-node20`

52075 passed, 1241 skipped

---

Merge workflow run.

[github-actions]

Test results for "MCP"

6 failed
❌ [chrome] › mcp/cli-webmcp.spec.ts:162 › webmcp-call disambiguates duplicate tool names by frame @mcp-macos-latest-chrome
❌ [firefox] › mcp/annotate.spec.ts:291 › should enter annotate mode on fresh dashboard.tsx mount with -s --annotate @mcp-windows-latest-firefox
❌ [firefox] › mcp/cli-devtools.spec.ts:140 › request* and response* commands support --filename @mcp-windows-latest-firefox
❌ [firefox] › mcp/cli-devtools.spec.ts:201 › recording-start-stop @mcp-windows-latest-firefox
❌ [firefox] › mcp/cli-mouse.spec.ts:38 › mousewheel @mcp-windows-latest-firefox
❌ [firefox] › mcp/cli-parsing.spec.ts:88 › should preserve leading zeros in string arguments @mcp-windows-latest-firefox

8699 passed, 1474 skipped

---

Merge workflow run.

[github-actions]

🟡 Nothing here points at this PR, but one failure I can't fully clear

Hi, I'm the Playwright bot and I took a look at the red CI.

This PR is a WebKit binary roll — browsers.json and webkit/protocol.d.ts, nothing else. All 6 failures are on chrome and firefox, and WebKit itself is green. Five of the six are tests with a documented history of flaking on unrelated SHAs. The sixth has no prior failure on record, so I'm leaving it uncertain rather than calling it a flake.

Details

Main suite (tests 1) is clean: 52075 passed, 5 flaky, 0 failed. Everything below is from the MCP report, run 35847214605. Five of the six failures land on the single mcp-windows-latest-firefox shard, which carries at least one failure in 319 of 702 recorded runs and three or more in 49 of them — a multi-failure run on that bot is routine, not a signal.

This PR's run isn't in the aggregated results DB yet, so every run cited below is a different SHA.

Pre-existing flake / infra

• [firefox] › mcp/annotate.spec.ts:291 › should enter annotate mode on fresh dashboard.tsx mount with -s --annotate — flips verdict regularly on mcp-windows-latest-firefox: failed in runs 31731606614 (chore(net): use native happy eyeballs instead of the manual implementation #42240), 32907115201 (fix(trace-viewer): store snapshot style text in an attribute #42405), 32979149779 (docs(browsercontext): undeprecate setHTTPCredentials #42419), 33216547259 (fix(cli): stop daemon when socket disappears #42454), 34387431006 (fix(screencast): keep overlays above top layer elements #42642) and two pushes, passing the other ~700. It also flakes on chrome, chromium, msedge and webkit, so it isn't engine-specific.
• [firefox] › mcp/cli-mouse.spec.ts:38 › mousewheel — same picture, 5 failures across ~720 runs, all on this bot: 31792089467 (chore(expect): pass expect title explicitly to library calls #42250), 32318630304 (feat(codegen): add --http-credentials option #42322), 32979149779 (docs(browsercontext): undeprecate setHTTPCredentials #42419), 34915545227 (Fix spelling of restoring in credentials docs #42718) and a push. Never failed on chromium, chrome, msedge or webkit.
• [firefox] › mcp/cli-devtools.spec.ts:201 › recording-start-stop — failed in 33089818898 (feat(reporter): introduce TestStep.subtitle, unify step formatting #42422), 33790327052 (devops: restore npm publishing from GitHub Actions #42550) and 35643037767 (a push two days ago), green in the other ~530.
• [firefox] › mcp/cli-parsing.spec.ts:88 › should preserve leading zeros in string arguments — failed in 34218129372 (feat(mcp): expose page-registered WebMCP tools #42613), 35592445344 (fix typo Provisional instead of Provisiolal #42829) and a push, green in the other ~720. A pure string-parsing test, so a WebKit revision bump can't reach it.
• [chrome] › mcp/cli-webmcp.spec.ts:162 › webmcp-call disambiguates duplicate tool names by frame — the newest and flakiest of the group: 3 failures in 96 recorded runs, on 35415860576 (fix(network): resolve response.finished() when the request fails after the response #42787), 35722584824 (fix(mcp): rebind page-registered WebMCP tools when the tab or frame changes #42821) and a push 35754542981. Also flaky on chromium. Worth a separate look on its own merits, but unrelated to this roll.

Uncertain

• [firefox] › mcp/cli-devtools.spec.ts:140 › request* and response* commands support --filename — this one has passed 724 of 724 recorded runs on firefox and has never failed on any project, so I can't call it a flake on the evidence. What makes me doubt it's real: a WebKit binary revision has no path to a Firefox MCP test, and it failed in the same shard as four proven flakes above. To settle it, a rerun of the mcp-windows-latest-firefox job — if it goes green there, that's a first data point for the flake column.

_{Triaged by the Playwright bot - agent run}

[github-actions]

Test results for "tests 2"

2 fatal errors, not part of any test
3 failed
❌ [chromium-library] › library/chromium/chromium.spec.ts:436 › should produce network events, routing, and annotations for Service Worker (advanced) @chrome-macos-latest
❌ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:470 › top level 'Partitioned;' cookie and same origin iframe @firefox-macos-15-xlarge
❌ [android-page] › page/page-set-content.spec.ts:57 › should include shadow roots

25 flaky

⚠️ [chromium-library] › library/video.spec.ts:762 › screencast › should work with video+trace `@chromium-windows-latest`
⚠️ [chromium-library] › library/chromium/chromium.spec.ts:179 › serviceWorker(), and fromServiceWorker() work `@chromium-macos-15-large`
⚠️ [chromium-library] › library/chromium/chromium.spec.ts:436 › should produce network events, routing, and annotations for Service Worker (advanced) `@chromium-macos-15-large`
⚠️ [chromium-library] › library/chromium/chromium.spec.ts:301 › should report intercepted service worker requests in HAR `@chromium-macos-14-xlarge`
⚠️ [chromium-library] › library/chromium/oopif.spec.ts:179 › should take screenshot `@chromium-macos-14-xlarge`
⚠️ [chromium-library] › library/video.spec.ts:762 › screencast › should work with video+trace `@chromium-macos-14-xlarge`
⚠️ [chromium-library] › library/chromium/oopif.spec.ts:179 › should take screenshot `@chromium-macos-15-xlarge`
⚠️ [chromium-library] › library/chromium/chromium.spec.ts:436 › should produce network events, routing, and annotations for Service Worker (advanced) `@chrome-beta-ubuntu-22.04`
⚠️ [chromium-library] › library/inspector/cli-codegen-2.spec.ts:202 › cli codegen › should upload a file via hidden input triggered by button `@chrome-beta-ubuntu-22.04`
⚠️ [chromium-library] › library/selector-generator.spec.ts:457 › selector generator › should match in shadow dom `@chrome-macos-latest`
⚠️ [chromium-library] › library/selector-generator.spec.ts:703 › selector generator › should generate noText: description from external aria-describedby `@chrome-macos-latest`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:257 › third party 'Partitioned;' cookies `@tracing-firefox`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:470 › top level 'Partitioned;' cookie and same origin iframe `@tracing-firefox`
⚠️ [firefox-page] › page/page-wait-for-function.spec.ts:43 › should poll on interval `@tracing-firefox`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:257 › third party 'Partitioned;' cookies `@firefox-macos-15-large`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:470 › top level 'Partitioned;' cookie and same origin iframe `@firefox-macos-15-large`
⚠️ [firefox-library] › library/defaultbrowsercontext-1.spec.ts:89 › should support viewport option `@firefox-macos-15-large`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:257 › third party 'Partitioned;' cookies `@firefox-windows-latest`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:470 › top level 'Partitioned;' cookie and same origin iframe `@firefox-windows-latest`
⚠️ [firefox-library] › library/signals.spec.ts:25 › should close the browser when the node process closes `@firefox-windows-latest`
⚠️ [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:257 › third party 'Partitioned;' cookies `@firefox-macos-15-xlarge`
⚠️ [firefox-library] › library/trace-viewer.spec.ts:1714 › should highlight locator in iframe while typing `@firefox-macos-15-xlarge`
⚠️ [webkit-library] › library/trace-viewer.spec.ts:2006 › canvas disabled title `@webkit-macos-15-xlarge`
⚠️ [webkit-library] › library/tracing.spec.ts:533 › should produce screencast frames scale `@webkit-macos-15-xlarge`
⚠️ [webkit-library] › library/screencast.spec.ts:57 › applies backpressure while async onFrame callback is pending `@webkit-macos-15-large`

112067 passed, 4744 skipped, 29 did not run

---

Merge workflow run.

[github-actions]

🟢 None of these failures come from this PR

Hi, I'm the Playwright bot and I took a look at the red CI.

This roll changes the WebKit revision 2365 → 2366 and a doc comment in webkit/protocol.d.ts — nothing else. The three failures in the latest report are on Chromium, Firefox and Android, and WebKit itself is green. All three have a failure record on SHAs this PR can't be responsible for; the Android one is currently broken on main, including this PR's own base commit.

Details

Latest report is tests 2, run 35853563796 — 112067 passed, 3 failed, 25 flaky. tests 1 is clean (0 failed) and MCP was triaged separately.

Pre-existing flake / infra

• [chromium-library] › library/chromium/chromium.spec.ts:436 › should produce network events, routing, and annotations for Service Worker (advanced) @chrome-macos-latest — fails with Test timeout of 30000ms exceeded. Error: request.response: Worker closed, 3 of 341 runs on this bot, green on the other 17 Chromium bots (~6800 runs, zero failures). Same test, same bot, same error on push bcb3563a and on feat(webkit): roll to r2358 #42438 (run 33129340741). It also flaked-and-recovered three more times inside this very report (chromium-macos-15-large, chrome-beta-ubuntu-22.04).

• [firefox-library] › library/browsercontext-cookies-third-party.spec.ts:470 › top level 'Partitioned;' cookie and same origin iframe @firefox-macos-15-xlarge — the most established flake of the three: 14 failures in 311 runs on this bot, zero on the other four Firefox bots. Recent unrelated failures include pushes 2a095f6f, bf9d04f2, 500c9c82, d78a9e71 and PR feat(firefox): roll to r1548 #42747 (run 35105295536) — always the same expect(locator).toHaveText on the iframe. Its sibling at :257 flaked on four separate bots in this report too.

• [android-page] › page/page-set-content.spec.ts:57 › should include shadow roots — not a flake, a real pre-existing breakage: page.content: TypeError: root.getHTML is not a function. It has failed 13 of 13 recorded runs on the android bot, all of them pushes to main since 2026-09-22, and never passed in the retained window. That includes push e125b2ff — which is exactly this PR's base commit, failing identically without the roll. The Android bot's WebView is too old for Element.getHTML(), so this needs a fix of its own; it's unrelated to a WebKit revision bump.

The 2 fatal errors and 29 did-not-run entries sit alongside the aborted Android shard, so I'd read them as the same story rather than a fourth one.

_{Triaged by the Playwright bot}

_{Triaged by the Playwright bot - agent run}