[teamleaderleo]

Base-red-job: web / react-apps-check
Base-red-job: web / react-apps-check (5)

Red

test/transcript-continuation-spacing.test.ts, "the same assistant's next reply and another agent's line take the block gap; a prompt keeps the turn gap", fails on base at expect(actual.peer).toBeCloseTo(gridLength("gap-block"), 1): Expected 24, Received 16 (base run 41425459815, job 125211076111, shard 5).

Culprit

#31879 ("Agent chat: a run of rows and status lines sit 16 from the prose around them"). Its second commit gave every status row (model.ts isStatusRow) MARK_STATUS, and transcriptGap(MARK_MESSAGE, MARK_STATUS) is the run gap (16). isStatusRow lists another agent's received message (kind === "user" with incoming.kind === "message"), which is exactly the gallery's continuation-peer row. #31946 had retargeted this test to the block gap 15 minutes before #31879 landed.

Is 16 intended?

Yes, and it is no overlap: the peer line's top sits 16px under the reply's bottom, a positive gap. The grid helper says so (transcriptGap(MARK_MESSAGE, MARK_STATUS) === TRANSCRIPT_RUN_GAP, asserted in continuationSpacing.test.ts), and the specs:

• design/transcript-spacing.md, Rules 3 and 4: received message lines are one-line activity rows, and "a run of rows is 16 from the prose around them"; The table, prose-to-step / step-to-prose.
• design/chat-transcript.md, Always rules: "another agent's received line" is a one-line activity row; "Rows between replies": another agent's message is an isStatusRow row.
• design/grid.md 5.5 still lists a status line under the 24 block gap; transcript-spacing.md (Agent chat: a run of rows and status lines sit 16 from the prose around them #31879) is the newer rule for runs and status lines.

Change

Test only. The peer assertion reads transcriptGap(MARK_MESSAGE, MARK_STATUS) instead of a literal token, as #31791 and #31946 did for the other gaps. No product CSS or transcriptGrid.ts change. The same-assistant gap (block gap), footer gap, turn gap and padding assertions are unchanged.

Evidence

Fast-checks runs at head 05bff4b6037cda74be0830e1b98cf97d040e3c1b, cd webviews && bun run test test/transcript-continuation-spacing.test.ts:

• Base's test restored: red, Expected: 24, Received: 16 on the peer gap (run 41429591328).
• This PR's test: green, 1 file, 0 failed (run 41429614955).
• Base: red in web / react-apps-check (5), job 125211076111 (run 41425459815).

Agent session: agent session

🤖 Generated with Claude Code

[cursor]

Bugbot needs on-demand usage enabled

Bugbot 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.

[github-actions]

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

Review 05bff4b: 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=pr32138 -f ref=lane/web-base-red-continuation-spacing -f platform=macos -f backend_mode=local

Shipped in:

• cmux NEXT nightly 1.0.0-nightly.4143649268801 (d887a9ec1b)
• cmux NEXT nightly 1.0.0-nightly.4143770410701 (45490d20ae)
• cmux NEXT nightly 1.0.0-nightly.4143839654801 (3a5b96c9c8)
• cmux NEXT nightly 1.0.0-nightly.4143934830301 (1e057e0cf1)
• cmux NEXT nightly 1.0.0-nightly.4144017965101 (18e36f549a)

[teamleaderleo]

Review (subagent claude-opus-5-5 medium): SHIP at 05bff4b

The 16 gap is meant, not swept in by accident. An incoming message has been an isStatusRow since #29124, where the doc comment names "another agent's message or notice". The second commit of #31879 deliberately gives every isStatusRow line MARK_STATUS and the run gap. The specs agree: design/transcript-spacing.md rules 3 and 4 put received lines with the run 16 from prose, and design/chat-transcript.md names "another agent's received line" as a one-line activity row. The fixture row continuation-peer is a user row with incoming.kind: "message".

The test still checks these against the live page: same reply to reply = gap-block, the hover glyphs clear the foot gap, footer = gap-footer, prompt = gap-turn, and zero paddings. The peer check is not vacuous: the expected value (16) differs from the old 24.

Follow-ups: design/grid.md 5.5 still says another agent's line takes the block gap. Update it to the run gap so it matches transcript-spacing.md rule 4.