[teamleaderleo]

Leo: "in that top section fade would be good"

Leo-ask: ask-cbdc0a8879 (sticky header fade, via claude-sr2)

What changed

An opened run of tool calls ("Read files, ran commands ⌄") lists its calls in a box that scrolls once it passes nine rows (.cv-tool-run). The summary line stays put above that box while the calls scroll under it. Before this change, a row scrolled halfway out was cut flat against the summary line: its top half was clipped, right under the header text.

Now the box's top edge fades out while there are rows above it:

• ToolRun.tsx attaches the existing useScrollEdges callback ref (summary/scrollEdges.ts) to the list. It sets data-fade-top only while the list is scrolled down from its start.
• conversation.css .cv-tool-run[data-fade-top] uses the same technique as the transcript's own top edge (.acpmux-scroll[data-fade-top], also 18px): an alpha mask on the scroller, transparent at the top and solid 18px below. Since nothing is painted over the rows, the fade can't cover the summary line's text or chevron, and it can't catch clicks. Because it is a mask and not a gradient, it works the same in light, dark and glass/translucent themes. The scrollbar lane on the right stays fully drawn. With Reduce Transparency the edge stays plain, as the transcript's does.
• No behaviour changes: same markup, same fold, same scrolling.

Builds on this lane's #24050 (useAnchoredCollapse). That PR changes how ToolRun opens and closes but not the list element, so the two merge cleanly in either order.

Before / after

Gallery state agent-pane.transcript / tool-run-scrolled-fade (synthetic fixtures): an 18-call run, opened and scrolled 3.5 rows so the fourth row sits half under the summary line.

• Before/after side by side (merge-base vs this branch): https://cmux-lawrences-mac-mini.tail137216.ts.net:18796/compare/tp-sticky-header-fade/?entry=agent-pane.transcript
• The state with only the gallery entry (no fade), for reference: https://cmux-lawrences-mac-mini.tail137216.ts.net:18796/compare/tp-sticky-header-fade-before/?entry=agent-pane.transcript

Test

webviews/test/tool-run-scroll-fade.test.tsx:

• When a run is opened, its list has no data-fade-top at the start. Scrolled 3.5 rows it has the mark, and scrolled back to the top the mark is gone.
• The fade rule is a mask (mask-image and -webkit-mask-image, transparent to #000 over 18px). It has no background and no pseudo-element over the rows, so it paints nothing and has nothing to catch the pointer. Reduce Transparency clears the mask.

Observed / Not run

• Observed: the web lint and pre-push gates passed on push. The gallery compare is requested on the fleet gallery host.
• Not run locally: bun test. cosmic-gray's hook refuses test runs, so the new test runs in CI (cmux-next-web).

No capture: webview-only change; the before/after is the Linux gallery above.

Changelog

• Fixed: In agent chat, an opened tool run's list now fades out under its summary line while scrolled, instead of cutting a row in half against it.

Agent session: cosmic-gray

🤖 Generated with Claude Code

---

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

---

Summary by cubic

Fades the top of an opened tool run's call list while it's scrolled, so a row half scrolled away dissolves under the summary line instead of sitting cut against it.

• The list marks its top edge via the existing useScrollEdges hook only while rows lie above it, and an 18px alpha mask (the transcript's own top-fade technique) fades that edge.
• Nothing is painted, so the fade covers no text and catches no clicks; it works in light, dark, and glass themes, while Reduce Transparency keeps the edge plain.
• A unit test covers the mark being set and cleared and the mask rule itself; a gallery fixture shows the run scrolled 3.5 rows.

^{Written for commit cdd36e5. Summary will update on new commits.}

[github-actions]

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

Review cdd36e5: 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=pr24263 -f ref=tp/sticky-header-fade -f platform=macos -f backend_mode=local

Gallery before/after for cdd36e5: 16 states rendered, 0 images differ, 661 more touched states not rendered. Images are in the run's artifact, not on this public PR.

gh run download 39875439925 -R manaflow-ai/cmux -n gallery-advisory-24263-cdd36e57e216c76a5ef1b4cd6198d9ab716fde95 (then read gallery-review.json)

States

State	Theme	Status	Diff	Picked by
agent-pane.chat-find-bar/empty	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.chat-find-bar/empty	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.chat-find-bar/matches	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.chat-find-bar/matches	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.chat-find-bar/next-match	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.chat-find-bar/next-match	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.chat-find-bar/no-results	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.chat-find-bar/no-results	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.diff-panel/last-turn	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.diff-panel/last-turn	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.diff-panel/toolbar	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.diff-panel/toolbar	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.diff-panel/tree-selection	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.diff-panel/tree-selection	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.diff-panel/open-file-error	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.diff-panel/open-file-error	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.diff-panel/escape	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.diff-panel/escape	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.diff-panel/review-file	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.diff-panel/review-file	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.diff-panel/stage-hunk	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.diff-panel/stage-hunk	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.diff-panel/more-lines	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.diff-panel/more-lines	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.path-chip/file	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.path-chip/file	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.path-chip/folder	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.path-chip/folder	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.path-chip/outside-lock	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.path-chip/outside-lock	Apple System Colors	unchanged	0%	own conversation.css
agent-pane.path-chip/outside-text	Apple System Colors Light	unchanged	0%	own conversation.css
agent-pane.path-chip/outside-text	Apple System Colors	unchanged	0%	own conversation.css

_{Pick other states with a Gallery: entry.id, entry.id/variant line in the PR body.}

red: cmux-next checks (god files, concurrency, crash safety, l10n); NEW: red check: Checks job structure (+2 more) is not red on the base 864ae80; do not merge until green; see lane-rules

Tour of cdd36e57e2 (run): no media. The tour did not run: only webview code changed (4 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, then re-add media-tour, to tour it on a Mac.

[blacksmith-sh]

Found 10 test failures on Blacksmith runners:

Failures

Test	View Logs
/tmp/tmpauhet2vo/summary-missing.json (live):/summary-missing.json (live):	View Logs
/tmp/tmpo8st9790/summary-fail.json (live):/summary-fail.json (live):	View Logs
[new_workspaces] debug.new_workspace timed 0 of 50 new workspace rows/ [new_workspaces] debug.new_workspace timed 0 of 50 new workspace rows	View Logs
[new_workspaces] the window draws no frames/[new_workspaces] the window draws no frames	View Logs
[tabs] debug.tab_switch timed 0 of 40 switches/ [tabs] debug.tab_switch timed 0 of 40 switches	View Logs
[tabs] the window draws no frames/[tabs] the window draws no frames	View Logs
test_base_passes (main.NoticesPairTests.test_base_passes)/ test_base_passes (main.NoticesPairTests.test_base_passes)	View Logs
test_every_submodule_free_checkout_on_an_owned_runner_resets_submodules (main.Reuse dWorkspaceSubmodules.test_every_submodule_free_checkout_on_an_owned_runner_resets_submo dules)/ test_every_submodule_free_checkout_on_an_owned_runner_resets_submodules (main.Reuse dWorkspaceSubmodules.test_every_submodule_free_checkout_on_an_owned_runner_resets_submo dules)	View Logs
test_every_submodule_free_checkout_on_an_owned_runner_resets_submodules (main.Reuse dWorkspaceSubmodules.test_every_submodule_free_checkout_on_an_owned_runner_resets_submo dules) (job='web-bundles')/ test_every_submodule_free_checkout_on_an_owned_runner_resets_submodules (main.Reuse dWorkspaceSubmodules.test_every_submodule_free_checkout_on_an_owned_runner_resets_submo dules) (job='web-bundles')	View Logs
test_targets_outside_the_app_go_to_the_mac_canary (main.SwiftCanaryFleetCoverage.te st_targets_outside_the_app_go_to_the_mac_canary)/ test_targets_outside_the_app_go_to_the_mac_canary (main.SwiftCanaryFleetCoverage.te st_targets_outside_the_app_go_to_the_mac_canary)	View Logs

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

[teamleaderleo]

cmux-next guards / nightly-guards / workflow-guard-tests / release-notary is red on 15 of 15 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.

[teamleaderleo]

Bead cx-io0nh created for this PR by the cmux coordinator.

[teamleaderleo]

Review (codex xhigh, correctness first): SHIP at 4fb0d31

• No correctness defects found. The stable callback ref attaches useScrollEdges to the actual list and disconnects its ResizeObserver, MutationObserver, and scroll listener when the ref is cleared; the fold animation keeps its separate parent ref.
• The alpha mask is limited to .cv-tool-run[data-fade-top], preserves the declared right-side scrollbar lane, paints no overlay, and is disabled for both Reduce Transparency paths. It does not cover the summary line or add a click target, and it is theme-independent.
• The gallery state and jsdom test cover opening, scrolling into the list, and returning to the top, plus the mask and transparency CSS claims. No chat-find or fold interaction defect is evident from this wiring.

[teamleaderleo]

Orphaned: no activity for 13 h, and no lane holds a lease on it. The orphan sweep could not merge feat-cmux-next in: conflicts in webviews/src/agent-session/acpmux/conversation/ToolRun.tsx. Listed for a lane to resolve.

[teamleaderleo]

Review (codex xhigh, correctness first): SHIP at cdd36e5

• No correctness defects found: the existing useScrollEdges callback ref is attached to the actual scroll container, and its lifecycle follows the fold mount/unmount path.
• The two mask layers keep the scrollbar lane opaque; the fade is alpha-only, has no overlay or pseudo-element, and is disabled by both Reduce Transparency paths.
• Fold animation, chat-find opening, light/dark/glass behavior, and the merged failure count remain intact.
• The jsdom test covers ref-driven edge toggling and the CSS contract; it cannot prove browser pixel rendering or observer teardown, but those are delegated to the existing hook and no defect is exposed by this diff.

[teamleaderleo]

cmux-next / cmux-next checks (god files, concurrency, crash safety, l10n) › Checks job structure is red on 9 of 14 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.

[github-actions]

Shipped in:

• cmux NEXT nightly 1.0.0-nightly.3991367032401 (b9c3d5aabc)
• cmux NEXT nightly 1.0.0-nightly.3991503232101 (cd84bc62d1)
• cmux NEXT nightly 1.0.0-nightly.3991575404801 (a62351daaa)