[saphid]

Problem

MessagesTimeline work rows > keeps the worked-for heading in place when expanded at the bottom (added in #29166) fails intermittently in CI. It failed on main itself (run 41478224580) and in unrelated PRs (for example run 41528296966):

TimeoutError: locator.click: Timeout 14526ms exceeded.
  - <div class="legend-list-content-container">…</div> intercepts pointer events

Legend List renders its rows with opacity: 0; pointer-events: none until it is ready, which is after its initial scroll settles. The test waited only for the scroll to reach the end. toBeVisible ignores opacity, so the test could read the heading's position and install fake timers while the rows were still hidden. With timers faked, Legend's readiness check never runs, so the click can never land. If readiness instead finishes later, it moves the heading, and the position assertion fails (expected 577 to be close to 473).

Change

The existing settle waitFor now also requires heading.element().checkVisibility({ opacityProperty: true }) before the test reads top and installs fake timers. This is a test-only change; the assertions the regression test makes stay the same.

Scope and approval

Very small, focused fix for an obvious bug, so no prior issue: a CI flake on main with an identified root cause, fixed in one test. Overlap search: no open PR or newer commit touches this test since #29166. #28125 (Legend List 3.7.1 upgrade) is a broad refactor and does not address this wait. The separate, more frequent AttachmentFilePreview PDF scroll/zoom flake is not part of this PR.

Verification

• Red on main: test with { repeats: 40 } under full CPU load (18 busy processes): 2 of 41 repeats failed, one with the click-intercept timeout and one with expected 577 to be close to 473. A temporary diagnostic captured, at the failing moment, the hit target legend-list-content-container and the rows' wrapper styled opacity: 0; ... pointer-events: none;.
• Green: same load, 3 rounds × 41 repeats, 123/123 passed.
• vp test run src/components/chat/MessagesTimeline.browser.test.tsx: 38/38 passed (rerun after rebasing on current main).
• vp run check:changed: format, lint, knip, typecheck, tests all pass.
• Not checked: the full web suite in CI (CI runs it).

Reviewed with sol-loop: 1 round with GPT-6.1 Sol on high. Sol verdict: approved.

Created with Claude Opus 5.5 in Claude Code (T3 Code).

🤖 Generated with Claude Code

[macroscopeapp]

Approvability

Verdict: Approved at c0dd3b4

Macroscope's review found this PR approvable — The change is confined to an ignored browser test and is described as waiting for timeline visibility before continuing. Its scope indicates no production behavior change or material risk requiring human judgment.

Notes:

• No code objects were reviewed. Approvability was decided on eligibility alone.

^{You can add or adjust custom eligibility rules. Learn more.}

[coderabbitai]

No actionable comments were generated in the recent review.

Recent review info

Run configuration

• Source mode: Auto
• Repository: pingdotgg/t3code
• Branch: main
• File: .coderabbit.config.ts
• Commit: 30084f0
• Layers: Repository TS config
• Review profile: CHILL
• Plan: Advanced
• Run ID: 377996d5-7ed1-4a83-b7c1-728547844a7e

Commits

Reviewing files that changed from the base of the PR and between 30084f0 and c0dd3b4.

Files selected for processing: 1

• apps/web/src/components/chat/MessagesTimeline.browser.test.tsx

Limit details: You’ve used all 10 included reviews currently available.

Walkthrough

The browser test now waits for the list to reach the bottom and for the expanded “Worked for” heading to become visible.

Changes: Worked-for heading test

Layer / File(s)	Summary
Synchronize heading visibility apps/web/src/components/chat/MessagesTimeline.browser.test.tsx	The test waits for the list to reach the bottom and checks heading visibility with opacity enabled.

Estimated code review effort: 1 (Trivial) | ~3 minutes

Change: Other

Suggested reviewers: yash-singh1

Merge Risk: Minimal · up to c0dd3

No actionable merge-blocking risk was identified; the synchronization change is ready to merge.

Pre-merge checks: 4

Check name	Status	Explanation
Title check		The title clearly identifies the test-only fix: the worked-for expansion test now waits for the timeline to finish loading.
Description check		The description includes Problem, Change, Scope and approval, and Verification sections. It explains the CI failure, root cause, fix, scope rationale, test results, and remaining verification limits.
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.

Finishing Touches

Generate unit tests (beta)

• Create a new PR

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Usage-based review receipt

• Mode: Continue automatically
• Reviewed files: 1
• Charged: $0.25
• View usage details

Note

This review was completed with usage-based billing: files reviewed beyond your plan's included limits are billed at $0.25/file. View usage-based billing.

[maria-rcks]

Note

Written by gpt-6.1-sol on behalf of Maria

Verified in a real dev environment

verified as-is at c0dd3b49708bf269253c46fbeb87245202adc101 on linux. this is a test-only readiness fix; production behavior is unchanged. no source changes or pushes.

• reproduced the original test failure on github/main (57cd05904d), in a separate worktree through blacksmith: expected 577 to be close to 473. the focused test failed without added cpu load or test modifications.
• on the exact pr head, all 38 tests in MessagesTimeline.browser.test.tsx passed through blacksmith. vp run check:changed 30084f0018c5751a07291222e8d04e16d513700a also passed format, lint, knip, affected typechecks, and the 38 timeline tests.
• ran the real dev web client with playwright and /usr/bin/google-chrome. created a project and threads through the app, then ran real codex gpt-6.1-sol turns through the local proxy. no replay, mocks, fixtures, or demo data.
• preserved the app-created conversation between main and pr. at the bottom, expanding worked-for kept the heading at y=433 on both revisions while scroll height grew from 843 to 2492, with scrollTop=195. collapse, scrolling to the final answer, expanding command details, and switching away and reopening the thread worked.
• a fresh thread on the pr completed a real provider turn; its command output returned the exact pr head sha.

main, expanded real turn:

pr, same conversation and viewport:

fresh pr provider turn and exact head output:

Not verified

native ios/android clients and macos/windows electron shells were not exercised on this linux vm. other providers and remote/relay/tunnel modes were not exercised. the test flake was reproduced once on main; the successful pr runs do not prove it can never recur.