Agent chat: the WebKit scroll-up test reads the offset after the wheel glide settles (base red) by teamleaderleo · Pull Request #23062 · manaflow-ai/cmux

teamleaderleo · 2026-10-21T06:29:11Z

Fixes the base red in webviews/test/agent-pane-stream-jitter.test.ts: "webkit: scrolling up while a reply streams is never pulled back".

Root cause (test timing, not the pane): Linux WebKit (WPE, which CI runs) animates a wheel step over several frames. The case read scrollTop 150ms after wheel(0, -400). On a loaded runner that read landed mid-glide, so the rest of the glide looked like a pull-back. A per-frame trace from CI on this PR's first head showed it:

the wheel at 605, then scrollTop 582 → 476 → 365 → 220 → 205 over ~330ms
the test read left as 220, 290, 322 or 500 depending on the run
then scrollTop held at 205 while the reply grew the content from 1201px to 1507px, with no pull-back toward the bottom

Every run ended at 205, the glide's end, so the pane never moved the reader. macOS WebKit and Chromium don't animate the wheel step here, which is why the case passes locally.

Fix: read the scrolled-up offset once scrollTop has held for 200ms (settledScrollTop), instead of after a fixed 150ms. The rest of the case is unchanged: 600ms after the turn ends the view must still be at that offset and more than 200px from the end, so a pane that re-pins or yanks the reader still fails.

There is no app change, so on base this case fails only where the wheel glides (CI's Linux WebKit). It passed 3/3 locally in Chromium and macOS WebKit.

Agent session: cmux16
🤖 Generated with Claude Code

^{Need help on this PR? Tag @codesmith-bot with what you need.}

…ic, temporary) The case is red on CI's Linux WebKit only (behind: true, kept: false) and passes on macOS WebKit under load; the trace shows what moved the offset. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

teamleaderleo · 2026-10-21T06:29:40Z

Owned by projects-57 on cmux16 · session

teamleaderleo · 2026-10-21T06:30:43Z

Bead cx-g0vgh created for this PR by the cmux coordinator.

github-actions · 2026-10-21T06:31:12Z

cmux status _{(one comment, edited in place; each line is updated by the job that owns it)}

Review 628ca9d: HQ feed card · gallery before/after runs · checks

App of this head (app.zip, kept 1 day)

gh workflow run reload-build.yml -R manaflow-ai/cmux --ref feat-cmux-next -f tag=pr23062 -f ref=scroll-jitter-2/webkit-scroll-up -f platform=macos -f backend_mode=local

Unclear: web / react-apps-check failed: no known signature; failed step: Check webview test shards. This PR merged before its CI finished: fix forward on main.

CI failed on 628ca9d76a (run 39559101710 attempt 1): 4 code, 1 unknown.

Job	Verdict	Why
web / react-apps-check (4)	code	a web test failed
web / react-apps-check (3)	code	a web test failed
web / react-apps-check (1)	code	a web test failed
web / react-apps-check (2)	code	a web test failed
web / react-apps-check	unknown	no known signature; failed step: `Check webview test shards`

Failures:

seen on other PRs agent-pane-wheel-pass-through.test.ts chromium > the wheel over a long draft stops at its ends and moves nothing else: failed (job)
seen on other PRs agent-pane-wheel-pass-through.test.ts webkit > the wheel over a long draft stops at its ends and moves nothing else: failed (job)
seen on other PRs chip-fixed-width.test.ts no ellipsized chip, tab or card in the agent pane truncates at a fixed width: failed (job)
seen on other PRs composer-history.test.ts ↑ and ↓ in an empty prompt step through the chat's earlier prompts: failed (job)
seen on other PRs diff-toolbar-narrow.test.ts at 460 px the base picker's label stays in its chip, clear of jump-to-file: failed (job)
seen on other PRs quiet-access-chip.test.ts chromium: Full access draws in the model chip's color: failed (job)
seen on other PRs quiet-access-chip.test.ts webkit: Full access draws in the model chip's color: failed (job)
seen on other PRs agent-pane-column-edges.test.ts the strip or queue and the card meet square in one rim, and the queue keeps room above: failed (job)
seen on other PRs agent-pane-no-capsules.test.ts agent pane page > rounds its controls as small-radius rects, not capsules: failed (job)
seen on other PRs composer-suggested-prompt.test.tsx nothing is suggested from another folder, another worktree, a merged PR, an old chat or a busy one: failed (job)
... 16 more in the job logs

Matched log lines

web / react-apps-check (4): (fail) chromium > the wheel over a long draft stops at its ends and moves nothing else [4078.84ms]
web / react-apps-check (3): (fail) the strip or queue and the card meet square in one rim, and the queue keeps room above [6356.15ms]
web / react-apps-check (1): (fail) with no running call, Ctrl-B stays the composer's and asks nothing [1121.19ms]
web / react-apps-check (2): (fail) queued and steering prompts > the queue takes its own place in the stack: nothing is cut off or drawn over [1394.56ms]

Not re-run automatically: web / react-apps-check (4), web / react-apps-check (3), web / react-apps-check (1), web / react-apps-check (2), web / react-apps-check are not machine failures.

Written by scripts/ci/classify_failures.py (ci-failure-attribution.yml); signatures are its SIGNATURES table. A machine verdict is the runner's fault, not this PR's; yours means the failing file is one this PR changes, also red on main that main's latest full suite fails the same way, seen on other PRs that it failed on another pull request's run lately.

red: web / Web status; NEW: green on the base c6b4a58; do not merge until green; see lane-rules
red: web / react-apps-check; no finished base run on c6b4a58 to compare; do not merge until green; see lane-rules
red: web / react-apps-check (1); no finished base run on c6b4a58 to compare; do not merge until green; see lane-rules
red: web / react-apps-check (2); no finished base run on c6b4a58 to compare; do not merge until green; see lane-rules
red: web / react-apps-check (3); no finished base run on c6b4a58 to compare; do not merge until green; see lane-rules
red: web / react-apps-check (4); no finished base run on c6b4a58 to compare; do not merge until green; see lane-rules

CI fast guards failed on a20a284377 (https://github.com/manaflow-ai/cmux/actions/runs/39555874619). It does not block the merge; a red guard merged into main breaks it for every open PR.
The log named no failed step; see the run.

Agents: python3 scripts/ci/guard_attribution.py fix applies the mechanical fixes locally.

Tour of 628ca9d76a (run): no media. The tour did not run: only webview code changed (1 path(s)), which the gallery advisory run checks in Linux Chromium on the pull request (its artifact holds any captures); add the media-tour-mac label to tour it on a Mac.

teamleaderleo · 2026-10-21T06:39:24Z

cmux-next guards / defaults-snapshot is red on 21 of 25 fresh feat-cmux-next PR heads in the last 2:00:00, so it is most likely a base red, not this PR's. The base-reds lane has it; a base merge picks up the fix once it lands.

blacksmith-sh · 2026-10-21T06:39:28Z

Found 22 test failures on Blacksmith runners:

Failures

Test	View Logs
`test/agent-pane-column-edges.test.ts/agent-pane-column-edges.test`	View Logs
`test/agent-pane-column-edges.test.ts/agent-pane-column-edges.test`	View Logs
`test/agent-pane-ctrl-b-background.test.ts/agent-pane-ctrl-b-background.test`	View Logs
`test/agent-pane-ctrl-b-background.test.ts/agent-pane-ctrl-b-background.test`	View Logs
`test/agent-pane-no-capsules.test.ts/agent-pane-no-capsules.test`	View Logs
`test/agent-pane-no-capsules.test.ts/agent-pane-no-capsules.test`	View Logs
`test/agent-pane-queue-steer.test.ts/agent-pane-queue-steer.test`	View Logs
`test/agent-pane-queue-steer.test.ts/agent-pane-queue-steer.test`	View Logs
`test/agent-pane-wheel-pass-through.test.ts/agent-pane-wheel-pass-through.test`	View Logs
`test/agent-pane-wheel-pass-through.test.ts/agent-pane-wheel-pass-through.test`	View Logs

...and 12 more test failures. View all on Blacksmith

^{Need help on this PR? Tag @codesmith-bot with what you need.}

…ttles Linux WebKit (WPE, as on CI) animates a wheel step over several frames. The case read scrollTop 150ms after the wheel, mid-glide on a loaded runner, and then saw the glide finish as a "pull back". A per-frame CI trace showed the view gliding 605 -> 205 and then holding at 205 while the reply grew by 300px: the pane never moved the reader. Read the offset once it has held for 200ms. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

github-actions · 2026-10-21T07:11:33Z

The feat-cmux-next push run https://github.com/manaflow-ai/cmux/actions/runs/39559424526 at 240de48 failed: cmux-next Release compile (Xcode 26).
Those jobs last passed at 233cc81. This pull request is one of 5 merged since: #23065, #23050, #23073, #22992, #23062.
If the failure is in your change, fix forward on feat-cmux-next. Pull requests run only the tiers their paths reach (docs/ci/cmux-next-tiers.md); label a batch PR full-ci to run them all before merging.

github-actions · 2026-10-21T11:24:27Z

Shipped in:

cmux NEXT nightly 1.0.0-nightly.3958372135301 (bc661dae0e)

teamleaderleo changed the title ~~Agent chat: scrolling up while a reply streams stays put in WebKit (base red)~~ Agent chat: the WebKit scroll-up test reads the offset after the wheel glide settles (base red) Oct 21, 2026

Merge branch 'feat-cmux-next' into scroll-jitter-2/webkit-scroll-up

628ca9d

teamleaderleo merged commit e1c2047 into feat-cmux-next Oct 21, 2026

teamleaderleo deleted the scroll-jitter-2/webkit-scroll-up branch October 21, 2026 07:01

github-actions Bot mentioned this pull request Oct 21, 2026

cmux-next: restyle the Extensions menu (site-access groups, two-line rows, accent pin, store link) #22992

Merged

teamleaderleo mentioned this pull request Oct 21, 2026

test(acpmux): column-edges lets a full queue scroll, so it holds on CI's fonts #23048

Merged

teamleaderleo mentioned this pull request Oct 22, 2026

Agent chat: a mouse wheel's notch glides the transcript instead of jumping #24327

Closed

Navigation Menu

Repository navigation

Agent chat: the WebKit scroll-up test reads the offset after the wheel glide settles (base red) - #23062

Agent chat: the WebKit scroll-up test reads the offset after the wheel glide settles (base red)#23062
teamleaderleo merged 3 commits into
feat-cmux-nextfrom
scroll-jitter-2/webkit-scroll-up

teamleaderleo commented Oct 21, 2026 •

edited by blacksmith-sh Bot

Loading

Uh oh!

teamleaderleo commented Oct 21, 2026 •

edited

Loading

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026 •

edited

Loading

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

blacksmith-sh Bot commented Oct 21, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

teamleaderleo commented Oct 21, 2026 • edited by blacksmith-sh Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

teamleaderleo commented Oct 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

blacksmith-sh Bot commented Oct 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Failures

Uh oh!

github-actions Bot commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

teamleaderleo commented Oct 21, 2026 •

edited by blacksmith-sh Bot

Loading

teamleaderleo commented Oct 21, 2026 •

edited

Loading

github-actions Bot commented Oct 21, 2026 •

edited

Loading

blacksmith-sh Bot commented Oct 21, 2026 •

edited

Loading