Repository navigation
Agent pane: status, tool and message rows and the action row line up with the prose - #29124
Merged
Merged
Conversation
Collaborator
Author
|
Owned by an agent session · session |
|
Found 1 test failure on Blacksmith runners: Failure
|
![Fix with [code]smith](https://pr-comments-assets.blacksmith.sh/codesmith/fix-with-codesmith-light-3x.png)
Contributor
|
cmux status (one comment, edited in place; each line is updated by the job that owns it) Review App of this head (app.zip, kept 1 day)
Gallery before/after for
States
Pick other states with a Passes: CI passes on CI passes on Written by web and size checks: green Shipped in:
|
teamleaderleo
force-pushed
the
lane/transcript-row-alignment
branch
from
October 31, 2026 10:11
4e0307f to
fa1b0a2
Compare
teamleaderleo
temporarily deployed
to
ci-routing-agent
October 31, 2026 10:11 — with 
GitHub Actions
Inactive
GitHub Actions
Inactive
teamleaderleo
force-pushed
the
lane/transcript-row-alignment
branch
from
October 31, 2026 10:25
fa1b0a2 to
869d573
Compare
teamleaderleo
temporarily deployed
to
ci-routing-agent
October 31, 2026 10:25 — with
GitHub Actions
Inactive
GitHub Actions
Inactive
teamleaderleo
temporarily deployed
to
ci-routing-agent
October 31, 2026 11:09 — with
GitHub Actions
Inactive
GitHub Actions
Inactive
… centred between replies A gallery entry (agent-pane.transcript-row-alignment) with a reply, a subagent that finished, a reply, a background command that ended, a reply, another agent's message and a reply, under a folded turn's Worked for line and a footer with its time, tokens and actions; and a browser test that measures every row's first ink against the prose's left edge, the iconed rows' shared text x, and the room above and below each status row. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?id=aid-92fa78a91e29868c5eaa
…ext inset A gallery state (agent-pane.agent-message-row / icon-column) stacks a command row, a message from a Codex session (its logo), this agent's send (the envelope) and a message from a sender with no mark (the bot). The browser test measures each top-level row's icon slot, glyph center, drawn glyph size and first word's x, and fails while a received message's mark sits 8px right of the tool rows' slot, its name 5.5px right of their words, and its logo draws at 14px beside the envelope's 11.5px. Leo: "the alignment i feel like we might as well" Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?id=aid-63b9964940c20354b639
Tool rows lay out as a grid: the 16px glyph slot at the row's left edge, the 8px gap, the words at 24px (grid.md 5.5). Messages did not: - a received message's line and sender each added 4px of inline padding, so its mark sat 8px right of the slot and its name started at 28px; the line drops the padding and the mark takes the slot (its hover chips keep their place before the chevron); - a sent message's line is flex, where the tool icon's -1.5px end margin (a no-op in the grid) pulled its words to 22.5px; it is 0 there; - the waiting group's chevron moves into the same slot. A harness logo draws at 12px, the stroke glyphs' drawn size, instead of 14px. Leo: "the alignment i feel like we might as well" Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?id=aid-63b9964940c20354b639
…t on the prose's left edge Leo, 2026-10-30: "see how claude code has those centered tool calls / id like more alignment too", and "1 CODEX 2 US 3 US 4 CODEX 5 CLAUDDE WE NEED TO LINE OUR SHIT UP LIKE THEIRS". - A status row between two replies (a tool run, a subagent's or a background command's end, another agent's message, a compaction, a failover, a session start: model.ts isStatusRow) has as much room under it as over it: the reply's action row (4 + 24) and the step's 2px margin, 30px each side (--cv-status-gap). It was 38 over and 4 to 10 under. Inside an opened Worked for, whose replies draw no action row, 12 each side. model.ts statusEdge marks the rows from the next drawn row alone (acpmux-row--before-status / --before-reply), and the estimates follow. - The reply's and the footer's action row starts its first glyph on the prose's left edge, its 24px button overhanging by the glyph's 4px inset; the footer's "Worked for 5s" label and the activity line under a plan drop their 2px inset. Rows that hold an action row keep layout containment alone, so the overhanging hover pill is not cut. - Gallery agent-pane.transcript-row-alignment gains roomy and off variants. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?id=aid-92fa78a91e29868c5eaa
teamleaderleo
force-pushed
the
lane/transcript-row-alignment
branch
from
October 31, 2026 11:33
d9e26de to
5179e65
Compare
teamleaderleo
temporarily deployed
to
ci-routing-agent
October 31, 2026 11:33 — with
GitHub Actions
Inactive
GitHub Actions
Inactive
teamleaderleo
temporarily deployed
to
ci-routing-agent
October 31, 2026 11:44 — with
GitHub Actions
Inactive
GitHub Actions
Inactive
teamleaderleo
added a commit
that referenced
this pull request
Oct 31, 2026
teamleaderleo
added a commit
that referenced
this pull request
Oct 31, 2026
teamleaderleo
added a commit
that referenced
this pull request
Oct 31, 2026
teamleaderleo
added a commit
that referenced
this pull request
Oct 31, 2026
teamleaderleo
added a commit
that referenced
this pull request
Oct 31, 2026
teamleaderleo
added a commit
that referenced
this pull request
Oct 31, 2026
teamleaderleo
added a commit
that referenced
this pull request
Oct 31, 2026
teamleaderleo
added a commit
that referenced
this pull request
Oct 31, 2026
teamleaderleo
marked this pull request as ready for review
October 31, 2026 12:51
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
teamleaderleo
temporarily deployed
to
ci-routing-agent
October 31, 2026 13:24 — with
GitHub Actions
Inactive
GitHub Actions
Inactive
#28018 made .cv-tool a flex row; the icon's -1.5px right margin, which the grid's fixed 16px track used to absorb, now pulled the label to 22.5px, off the agent-message rows' 24px. Drop it so every iconed row's words share one x (design/grid.md 5.5). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
teamleaderleo
temporarily deployed
to
ci-routing-agent
October 31, 2026 13:37 — with
GitHub Actions
Inactive
GitHub Actions
Inactive
check:design-tokens refuses a raw calc in padding-bottom; the derived value lives in --cv-status-gap-plain-over beside its parts. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Toward base: #29754's one icon column (--cv-row-icon, --cv-row-icon-gap, the tool text inset) is kept whole, and #29740 removed per-reply action rows (hasReplyActions is false), so base's reply-actions test stands. statusEdge now reads the room under a status row from the reply above, as the room over it, so the two always match (12px each side while replies draw no action row; #29585, #29853). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
teamleaderleo
temporarily deployed
to
ci-routing-agent
November 1, 2026 00:42 — with
GitHub Actions
Inactive
GitHub Actions
Inactive
teamleaderleo
temporarily deployed
to
ci-routing-agent
November 1, 2026 00:46 — with
GitHub Actions
Inactive
GitHub Actions
Inactive
Collaborator
Author
|
Awareness note from an agent lane that does not own this PR; it is not a review or a request to take the PR over. For awareness, at head c839b6f: the base merge keeps |
Collaborator
Author
|
Review (codex, router-review): SHIP at c839b6f Reviewer: codex gpt-6-astra high, fast tier. Follow-ups: 1. Each lands as an issue in this repo that links this receipt; the lane that opened the PR fixes it forward. SHIP at c839b6f Visual verdict: inspected all supplied after renders; no clipping, bloated controls, or off-screen layout.
Follow-ups:
|
Collaborator
Author
|
Review (independent Codex gpt-6.1-sol, high, via the review router; full diff from merge base db63437 to this head): BLOCK at c839b6f. [P2] [blocking] webviews/test/transcript-row-alignment.test.ts:224 - Artifact-enabled runs always time out here: |
Collaborator
Author
|
Review (independent Codex gpt-6.1-sol, high, via the review router; full diff from merge base db63437 to this head): BLOCK at c839b6f. [P2] [blocking] webviews/test/transcript-row-alignment.test.ts:224 - With |
…ere one exists #29740 left a turn's copy to its footer, so no reply in the gallery draws its own action row and the hover waited out its timeout. The geometry (every lead at 0, every status gap 12/12) already held. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Bugbot needs on-demand usage enabledBugbot uses usage-based billing for this team and requires on-demand usage to be enabled. A team admin can enable on-demand usage in the Cursor dashboard. |
teamleaderleo
temporarily deployed
to
ci-routing-agent
November 1, 2026 01:39 — with
GitHub Actions
Inactive
GitHub Actions
Inactive
teamleaderleo
added a commit
that referenced
this pull request
Nov 1, 2026
teamleaderleo
added a commit
that referenced
this pull request
Nov 1, 2026
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
There was a problem hiding this comment.
13 issues found across 13 files
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="webviews/test/transcript-icon-column.test.ts">
<violation number="1" location="webviews/test/transcript-icon-column.test.ts:32">
P2: This catch can silently skip the test in GitHub Actions after `requireBrowserLane` has already accepted the lane, allowing a transient Chromium launch failure to produce a green run with no alignment assertions. Use `probeEngineOrSkip("transcript-icon-column", "chromium", chromium)` here so required lanes fail while local runs retain the intended skip behavior.</violation>
</file>
<file name="webviews/src/agent-session/acpmux/conversation/MessageActions.tsx">
<violation number="1" location="webviews/src/agent-session/acpmux/conversation/MessageActions.tsx:337">
P2: The new `data-transcript-gap="blank"` marker turns the reply action row's empty box into a drag start, but no browser test covers it. Add a case that presses the blank box (and a button inside it) to confirm selection starts only on the box.</violation>
</file>
<file name="webviews/test/transcript-row-alignment.test.ts">
<violation number="1" location="webviews/test/transcript-row-alignment.test.ts:40">
P2: This catch can silently skip the suite on the required GitHub Actions browser lane. Let the launch error fail there, or use the shared browser probe so missing engines are skipped only in environments where skipping is permitted.</violation>
<violation number="2" location="webviews/test/transcript-row-alignment.test.ts:240">
P2: This test never requires or measures a reply action row. Hover a reply before `measure(page)` and require `"reply actions"` in `kinds`, otherwise regressions in that alignment can pass unnoticed.</violation>
</file>
<file name="webviews/src/agent-session/acpmux/model.ts">
<violation number="1" location="webviews/src/agent-session/acpmux/model.ts:1173">
P2: `isToolRun` treats a pending edit as a tool line, but the height estimate in `fallbackRowHeight` sizes the same row as an edited-files card. Check `isEditRow` and align the two rules so a pending-edit row is classed the same way in both places.</violation>
<violation number="2" location="webviews/src/agent-session/acpmux/model.ts:1195">
P2: `statusEdge` marks every assistant immediately before a status row, including terminal status rows that are not between replies. Require a later drawn assistant before applying the centered-status edge, and handle consecutive status rows as one status block.</violation>
</file>
<file name="webviews/src/agent-session/acpmux/App.tsx">
<violation number="1" location="webviews/src/agent-session/acpmux/App.tsx:3359">
P2: The kept-prefix reuse in the `measured` memo does not account for a status edge change on the row above the first changed row. The existing `hasReplyActions` check resets `kept` to that reply for this reason, but the new status edge has no equivalent, so an unmounted reply keeps its stale height until it mounts. Apply the same reset for `statusEdge` on the last drawn row before `kept`.</violation>
</file>
<file name="webviews/src/agent-session/acpmux/conversation/conversation.css">
<violation number="1" location="webviews/src/agent-session/acpmux/conversation/conversation.css:2585">
P3: This comment drops the 1.5px overlap, so it now describes a 24px inset. The rule uses `--cv-tool-text-inset`, which is 22.5px. Restore the overlap wording.</violation>
<violation number="2" location="webviews/src/agent-session/acpmux/conversation/conversation.css:2818">
P3: The new comment says "No inline padding", but this rule still sets `padding-inline: var(--ds-space-1)`. The negative margin cancels that padding so the mark lands on the column edge. Reword the comment to say that, and fold it into the older comment above the rule.</violation>
<violation number="3" location="webviews/src/agent-session/acpmux/conversation/conversation.css:2860">
P3: The hover pill now ends 16px from the line's end, which runs into the 20px chevron box instead of clearing it. The comment's "one small gap" is not true. Restore the previous 20px inset, or measure the chevron box before changing it.</violation>
<violation number="4" location="webviews/src/agent-session/acpmux/conversation/conversation.css:2886">
P3: The comment's 24px is wrong. The sender's name starts at 22.5px (16px slot plus 6.5px gap), matching `--cv-tool-text-inset`. Correct the number to 22.5px.</violation>
</file>
<file name="webviews/src/agent-session/acpmux/conversation/TranscriptRowAlignment.gallery.ts">
<violation number="1" location="webviews/src/agent-session/acpmux/conversation/TranscriptRowAlignment.gallery.ts:57">
P2: This fixture does not render the completed-subagent status row it claims to test: `SUBAGENTS` renders a group header, while `SubagentDoneLine` requires the `subagentDone` row kind. Use a `subagentDone` row here so the alignment scenario covers the intended line.</violation>
</file>
<file name="webviews/src/agent-session/acpmux/conversation/PendingMessagesCard.css">
<violation number="1" location="webviews/src/agent-session/acpmux/conversation/PendingMessagesCard.css:49">
P2: The waiting-message text still starts 1.5px to the right of tool and received text: its 8px flex gap omits the shared icon-overlap adjustment. Apply the tool row's negative icon margin to `.cv-pending__icon`.</violation>
</file>
Reply with feedback, questions, or to request a fix.
View guided diff | Turn on auto-fix | Re-trigger cubic
|
|
||
| try { | ||
| await (await chromium.launch({ headless: true })).close(); | ||
| } catch { |
There was a problem hiding this comment.
P2: This catch can silently skip the test in GitHub Actions after requireBrowserLane has already accepted the lane, allowing a transient Chromium launch failure to produce a green run with no alignment assertions. Use probeEngineOrSkip("transcript-icon-column", "chromium", chromium) here so required lanes fail while local runs retain the intended skip behavior.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/test/transcript-icon-column.test.ts, line 32:
<comment>This catch can silently skip the test in GitHub Actions after `requireBrowserLane` has already accepted the lane, allowing a transient Chromium launch failure to produce a green run with no alignment assertions. Use `probeEngineOrSkip("transcript-icon-column", "chromium", chromium)` here so required lanes fail while local runs retain the intended skip behavior.</comment>
<file context>
@@ -0,0 +1,204 @@
+
+ try {
+ await (await chromium.launch({ headless: true })).close();
+ } catch {
+ describe.skip("transcript icon column", () => test("browser", () => {}));
+ return;
</file context>
| // Over a status row this line is the reply's whole bottom room (conversation.css | ||
| // `--cv-status-gap`), so its own blank box, never its buttons, is a transcript gap a drag | ||
| // can start from (gutterSelect.ts). | ||
| <div className="cv-turn-actions cv-reply-actions" data-transcript-gap="blank"> |
There was a problem hiding this comment.
P2: The new data-transcript-gap="blank" marker turns the reply action row's empty box into a drag start, but no browser test covers it. Add a case that presses the blank box (and a button inside it) to confirm selection starts only on the box.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/src/agent-session/acpmux/conversation/MessageActions.tsx, line 337:
<comment>The new `data-transcript-gap="blank"` marker turns the reply action row's empty box into a drag start, but no browser test covers it. Add a case that presses the blank box (and a button inside it) to confirm selection starts only on the box.</comment>
<file context>
@@ -331,7 +331,10 @@ export function ReceivedActions({ row, more }: { row: AcpmuxRow; more?: ReactNod
+ // Over a status row this line is the reply's whole bottom room (conversation.css
+ // `--cv-status-gap`), so its own blank box, never its buttons, is a transcript gap a drag
+ // can start from (gutterSelect.ts).
+ <div className="cv-turn-actions cv-reply-actions" data-transcript-gap="blank">
<span className="cv-turn-action-group">
<MessageCopy text={text} />
</file context>
| "reply", | ||
| "worked-for", | ||
| "footer label", | ||
| "footer actions", |
There was a problem hiding this comment.
P2: This test never requires or measures a reply action row. Hover a reply before measure(page) and require "reply actions" in kinds, otherwise regressions in that alignment can pass unnoticed.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/test/transcript-row-alignment.test.ts, line 240:
<comment>This test never requires or measures a reply action row. Hover a reply before `measure(page)` and require `"reply actions"` in `kinds`, otherwise regressions in that alignment can pass unnoticed.</comment>
<file context>
@@ -0,0 +1,258 @@
+ "reply",
+ "worked-for",
+ "footer label",
+ "footer actions",
+ "tool",
+ "subagent finished",
</file context>
|
|
||
| try { | ||
| await (await chromium.launch({ headless: true })).close(); | ||
| } catch { |
There was a problem hiding this comment.
P2: This catch can silently skip the suite on the required GitHub Actions browser lane. Let the launch error fail there, or use the shared browser probe so missing engines are skipped only in environments where skipping is permitted.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/test/transcript-row-alignment.test.ts, line 40:
<comment>This catch can silently skip the suite on the required GitHub Actions browser lane. Let the launch error fail there, or use the shared browser probe so missing engines are skipped only in environments where skipping is permitted.</comment>
<file context>
@@ -0,0 +1,258 @@
+
+ try {
+ await (await chromium.launch({ headless: true })).close();
+ } catch {
+ describe.skip("transcript row alignment", () => test("browser", () => {}));
+ return;
</file context>
| for (const item of row.items ?? []) { | ||
| if (!item.tool) continue; | ||
| tool = true; | ||
| if (card && (item.tool.kind === "edit" || item.tool.kind === "fileChange") && item.tool.status !== "pending") |
There was a problem hiding this comment.
P2: isToolRun treats a pending edit as a tool line, but the height estimate in fallbackRowHeight sizes the same row as an edited-files card. Check isEditRow and align the two rules so a pending-edit row is classed the same way in both places.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/src/agent-session/acpmux/model.ts, line 1173:
<comment>`isToolRun` treats a pending edit as a tool line, but the height estimate in `fallbackRowHeight` sizes the same row as an edited-files card. Check `isEditRow` and align the two rules so a pending-edit row is classed the same way in both places.</comment>
<file context>
@@ -1144,6 +1144,81 @@ export const hasReplyActions = (_row: AcpmuxRow, _next: AcpmuxRow | undefined) =
+ for (const item of row.items ?? []) {
+ if (!item.tool) continue;
+ tool = true;
+ if (card && (item.tool.kind === "edit" || item.tool.kind === "fileChange") && item.tool.status !== "pending")
+ return false;
+ }
</file context>
Suggested change
| if (card && (item.tool.kind === "edit" || item.tool.kind === "fileChange") && item.tool.status !== "pending") | |
| if (card && (item.tool.kind === "edit" || item.tool.kind === "fileChange")) |
| align-items: center; | ||
| justify-content: center; | ||
| width: 16px; | ||
| width: var(--cv-glyph-slot); |
There was a problem hiding this comment.
P2: The waiting-message text still starts 1.5px to the right of tool and received text: its 8px flex gap omits the shared icon-overlap adjustment. Apply the tool row's negative icon margin to .cv-pending__icon.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/src/agent-session/acpmux/conversation/PendingMessagesCard.css, line 49:
<comment>The waiting-message text still starts 1.5px to the right of tool and received text: its 8px flex gap omits the shared icon-overlap adjustment. Apply the tool row's negative icon margin to `.cv-pending__icon`.</comment>
<file context>
@@ -43,7 +46,7 @@
align-items: center;
justify-content: center;
- width: 16px;
+ width: var(--cv-glyph-slot);
color: var(--cv-text-muted);
}
</file context>
Suggested change
| width: var(--cv-glyph-slot); | |
| width: var(--cv-glyph-slot); | |
| margin-right: calc(var(--cv-row-icon-gap) - var(--cv-tool-gap)); |
| ui is all quite inconsistent"): one plate under the whole line on hover, its name, preview, | ||
| chevron and actions in it; the preview takes the line's room, so every line's chevron ends at the | ||
| same place; the actions show only on hover or focus, over the preview's tail, never moving it. */ | ||
| /* No inline padding: the sender's mark sits in the tool rows' glyph slot at the row's left edge and |
There was a problem hiding this comment.
P3: The new comment says "No inline padding", but this rule still sets padding-inline: var(--ds-space-1). The negative margin cancels that padding so the mark lands on the column edge. Reword the comment to say that, and fold it into the older comment above the rule.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/src/agent-session/acpmux/conversation/conversation.css, line 2818:
<comment>The new comment says "No inline padding", but this rule still sets `padding-inline: var(--ds-space-1)`. The negative margin cancels that padding so the mark lands on the column edge. Reword the comment to say that, and fold it into the older comment above the rule.</comment>
<file context>
@@ -2791,6 +2815,10 @@ pre.cv-math-source {
ui is all quite inconsistent"): one plate under the whole line on hover, its name, preview,
chevron and actions in it; the preview takes the line's room, so every line's chevron ends at the
same place; the actions show only on hover or focus, over the preview's tail, never moving it. */
+/* No inline padding: the sender's mark sits in the tool rows' glyph slot at the row's left edge and
+ the chevron ends at its right edge, as every tool row's do (design/chat-transcript.md "One icon
+ column"; Leo, 2026-10-31: "the alignment i feel like we might as well"). The hover plate is the
</file context>
| } | ||
| /* A failed call's cause (ToolFailureCause.tsx, the `tool-error-look` experiment): one quiet line | ||
| under the row, starting under its words (the 16px glyph, less its 1.5px overlap, and the 8px | ||
| under the row, starting under its words (the 16px glyph and the 8px |
There was a problem hiding this comment.
P3: This comment drops the 1.5px overlap, so it now describes a 24px inset. The rule uses --cv-tool-text-inset, which is 22.5px. Restore the overlap wording.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/src/agent-session/acpmux/conversation/conversation.css, line 2585:
<comment>This comment drops the 1.5px overlap, so it now describes a 24px inset. The rule uses `--cv-tool-text-inset`, which is 22.5px. Restore the overlap wording.</comment>
<file context>
@@ -2558,7 +2582,7 @@ pre.cv-math-source {
}
/* A failed call's cause (ToolFailureCause.tsx, the `tool-error-look` experiment): one quiet line
- under the row, starting under its words (the 16px glyph, less its 1.5px overlap, and the 8px
+ under the row, starting under its words (the 16px glyph and the 8px
gap), cut to the width; the opened call has the rest. The error tone stays on the row's word. */
.cv-tool-error {
</file context>
Suggested change
| under the row, starting under its words (the 16px glyph and the 8px | |
| under the row, starting under its words (the 16px glyph, less its 1.5px overlap, and the 8px |
| name, which opens its chat. The name's hover highlight is its own, never the row's. The mark takes | ||
| the tool rows' glyph slot (--cv-glyph-slot wide at the row's left edge, glyph centered, then the | ||
| 8px gap), so a sender's logo and a send's envelope share one center line and the name starts where | ||
| every tool row's words do: the slot plus the gap, 24px (grid.md 5.5; transcript-icon-column.test.ts). */ |
There was a problem hiding this comment.
P3: The comment's 24px is wrong. The sender's name starts at 22.5px (16px slot plus 6.5px gap), matching --cv-tool-text-inset. Correct the number to 22.5px.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/src/agent-session/acpmux/conversation/conversation.css, line 2886:
<comment>The comment's 24px is wrong. The sender's name starts at 22.5px (16px slot plus 6.5px gap), matching `--cv-tool-text-inset`. Correct the number to 22.5px.</comment>
<file context>
@@ -2852,7 +2880,10 @@ pre.cv-math-source {
+ name, which opens its chat. The name's hover highlight is its own, never the row's. The mark takes
+ the tool rows' glyph slot (--cv-glyph-slot wide at the row's left edge, glyph centered, then the
+ 8px gap), so a sender's logo and a send's envelope share one center line and the name starts where
+ every tool row's words do: the slot plus the gap, 24px (grid.md 5.5; transcript-icon-column.test.ts). */
.cv-received__sender {
display: inline-flex;
</file context>
Suggested change
| every tool row's words do: the slot plus the gap, 24px (grid.md 5.5; transcript-icon-column.test.ts). */ | |
| every tool row's words do: the slot plus the gap, 22.5px (grid.md 5.5; transcript-icon-column.test.ts). */ |
| /* Before the chevron (12px) and the line's inset. */ | ||
| inset-inline-end: calc(var(--ds-space-1) + 12px + var(--ds-space-1)); | ||
| /* Before the chevron (12px), one small gap clear of it. */ | ||
| inset-inline-end: calc(12px + var(--ds-space-1)); |
There was a problem hiding this comment.
P3: The hover pill now ends 16px from the line's end, which runs into the 20px chevron box instead of clearing it. The comment's "one small gap" is not true. Restore the previous 20px inset, or measure the chevron box before changing it.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. At webviews/src/agent-session/acpmux/conversation/conversation.css, line 2860:
<comment>The hover pill now ends 16px from the line's end, which runs into the 20px chevron box instead of clearing it. The comment's "one small gap" is not true. Restore the previous 20px inset, or measure the chevron box before changing it.</comment>
<file context>
@@ -2828,8 +2856,8 @@ pre.cv-math-source {
- /* Before the chevron (12px) and the line's inset. */
- inset-inline-end: calc(var(--ds-space-1) + 12px + var(--ds-space-1));
+ /* Before the chevron (12px), one small gap clear of it. */
+ inset-inline-end: calc(12px + var(--ds-space-1));
display: inline-flex;
align-items: center;
</file context>
Collaborator
Author
|
Review (codex, router-review): SHIP at bc7616b SHIP at bc7616b Visual verdict: Light/dark after images show the intended alignment without clipped or repeated controls.
Follow-ups:
|
teamleaderleo
added a commit
that referenced
this pull request
Nov 1, 2026
Resolves #29124 (status rows centred between replies, shared icon column): App.tsx keeps its status edge class beside acpmux-row--line; the waiting group's head takes base's slot geometry and keeps no margins of its own; a one-line row's padding yields to the status edges (same specificity, later) and statusDelta counts a line row's 4px run gap as a tool run's. Also, from review: a message this agent sent another one (SendMessage, acpmux send, a sub-session's send_message) estimates as one 22px tool line. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?id=aid-013e821a578a652a3caa
This was referenced Nov 1, 2026
Merged
Merged
This branch was previously deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Leo, 2026-10-30: "see how claude code has those centered tool calls
id like more alignment too"
Leo, 2026-10-30: "1 CODEX 2 US 3 US 4 CODEX 5 CLAUDDE WE NEED TO LINE OUR STUFF UP LIKE THEIRS"
What changed
Agent pane transcript (cmux NEXT), alignment only. No row type, icon or action was removed, and the incoming agent-message action buttons were left alone.
statusEdge(model.ts) reads both sides from the reply above, and only a row with a reply on both sides takes the room: a turn that starts with a tool (prompt, tool, reply) keeps its own spacing. Since Transcript: copy once per turn and keep pre-steer replies visible #29740 replies draw no action row of their own (the turn's footer has Copy), so the room is the plain 12px each side (--cv-status-gap-plain). If a reply draws its own row again, both sides become--cv-status-gap(30px).model.tsestimates the same edge, so estimated heights match drawn ones.padding-inline-start: 2pxon.cv-turn-actionsand.cv-activityis gone, and the first 24px icon button overhangs the column by its glyph's 4px inset.--cv-row-icon,--cv-row-icon-gap) as merged. Agent pane: messages between agents share the tool rows' icon column and text inset #28796's two commits, carried here with their authorship, add the check that a message row's mark sits in the same 16px slot and draws at the same size as a tool row's. This PR subsumes Agent pane: messages between agents share the tool rows' icon column and text inset #28796.data-transcript-gap="blank", gutterSelect.ts), never from its buttons.Measured by the new browser test at 2x (both themes give the same numbers):
Before / after (2x, gallery
agent-pane.transcript-row-alignment)I opened both after images. Every row starts on the prose edge, and each status row sits in the middle of its gap.
Gallery variants:
mix(shipped),roomy(a wider gap),off(the old spacing), andfooter(the summary row hovered).Tests
webviews/test/transcript-row-alignment.test.ts(new, Chromium browser lane, light and dark) checks two things:webviews/test/transcript-icon-column.test.ts(from Agent pane: messages between agents share the tool rows' icon column and text inset #28796).rowEstimates.test.ts: the status-row estimate, and a status row under a prompt that keeps its own spacing.Risk
Layout only.
model.tsrow height estimates change for a status row next to a reply. The drawn height is reported with its status edge, and a mismatch drops the cached height, so a stale estimate is re-measured rather than kept.Performance
statusEdgeis O(1) per row and reads only the next row. It adds no timer, observer or animation loop, and no per-token work.Spec: design/chat-transcript.md, "Rows between replies" (leo-workspace#1916).
Agent session: agent session
🤖 Generated with Claude Code