[teamleaderleo]

Problem

The cmux NEXT agent chat hid most of a turn's work. When you opened a finished turn's Worked for, the run between two assistant messages was still one line, "Used a tool, read a file, ran commands >". Each "Ran N commands" group was closed too, so seeing any command took two or three clicks. MCP calls showed their raw ids in the running line, for example "··· mcp__cmux-subsessions__start_session 18s".

Change

• An opened Worked for lists each group and call on its own line, the way the Codex and Claude Code desktop apps do. The "Used tools, …" run fold now applies only when the Debug toggle codexToolRuns is on (foldRun, turns.ts codexRuns).
• Like the Codex app, command, read and search groups list their calls flush under the line, without the tree guide. Each command row reads "Ran " (the localized tools.ranCommand) after the terminal glyph.
• Command, read and search groups are open by default and show their last 5 calls. Earlier calls sit behind "+K more" (ToolGroupRow.tsx GROUP_TAIL), and Find opens that hidden part when it matches there. Edit groups stay closed because each one opens to a full diff. Your own open/close choices are still remembered per session.
• Default change (the defaults guard will ask for a call owner): the Debug tunable agentPane.debug.codexToolRuns now defaults to off (AgentPaneTunables.swift). The app sent true, so in nightly a running turn still folded each finished run into "Used tools, ran commands" and showed only the newest call of the run in progress, which is the behavior Leo flagged. The toggle stays available for comparison.
• A running turn shows its groups open as calls arrive, under "··· Working for …". A finished turn still collapses under "Worked for Xm Ys" with its divider. Neither of those changed.
• MCP calls are described in words:
  • Finished rows: start, launch and spawn tools read "Started session sidebar-drag-bugs" (new tools.mcp.start). Unknown verbs read "Called · ".
  • The running line, the activity line, ToolRow and the permission card no longer show mcp__server__tool. They show "cmux-subsessions · start session sidebar-drag-bugs" or "cmux · list workspaces" (mcpPlainName, mcpTitleName).
  • The raw id stays available as the row's tooltip and as the first line of the opened call (.cv-mcp-call__name).

Not in this PR: the Codex-style bordered sub-step cards, the background-agent status card and the pin footer action. The cmux-subsessions cards are being built by the subsession-visibility lane on top of this PR, as a new ToolGroup type, and fall back to these labels.

Screenshots (gallery agent-pane.transcript, before = feat-cmux-next 633c661, after = this branch)

Reference: the Codex desktop app's run (public screenshot from openai/codex#37213; that report's rows lost their command text, but the structure is what we match). It shows a "Ran commands ⌄" line, then one row per call: the terminal glyph and "Ran …", flush under the line. After this PR ours lists a group the same way. Each row reads "Ran " with the command in mono, and the exit status and time sit at the right edge.

Finished turn, Worked for opened (dense-ended-run)

Before	After

Running turn (dense-live-run)

Before	After

Verification

• New webviews/test/transcript-density.test.tsx (4 tests) and the updated TurnRows.test.tsx pass. Commit 1 adds the tests and commit 2 adds the fix.
• bun test over the transcript, agent-pane, thread and acpmux suites: 23 failures, the same 23 test names fail on base when files share a run. None are new.
• tsc --noEmit is clean. vp fmt was applied. The gallery tests (43) pass.
• Localization: two new keys, tools.more and tools.mcp.start, in all 21 locales of acpmux/Localizable.xcstrings. Non-English values are marked needs_review.

Agent session: cosmic-gray

🤖 Generated with Claude Code

[github-actions]

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

Review 8d57711: 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=pr22658 -f ref=leo/transcript-density -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 8d57711950 (run 39526533438 attempt 1): 1 code, 1 unknown.
macOS jobs did not run: compile and app tests were skipped: Fast static checks skipped.

Job	Verdict	Why
web / react-apps-check (1)	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-background-tasks.test.ts clicking a background task's end opens its terminal beside the chat: failed (job)

Matched log lines

web / react-apps-check (1): (fail) clicking a background task's end opens its terminal beside the chat [1058.05ms]

Not re-run automatically: web / react-apps-check (1), 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; base red: the same failure on the base df88ff8; does not block
red: web / react-apps-check; base red: the same failure on the base df88ff8; does not block
red: web / react-apps-check (1); base red: the same failure on the base df88ff8; does not block

CI fast guards failed on 643f1b35c7 (https://github.com/manaflow-ai/cmux/actions/runs/39524589064). 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 8d57711950 (run): no media. The tour job failed before the tour ran (build or setup); the run log has the output. Both apps run cmux-tui from 6d5a23f (tree 31bf86578b93): the merge base's own tree c7a23e61816b was not published yet.

[teamleaderleo]

Bead cx-42n1w created for this PR by the cmux coordinator.

[teamleaderleo]

Visual preview of 643f1b3: in the maintainers' private review feed.

[teamleaderleo]

cmux-next guards / defaults-snapshot is red on 20 of 20 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]

Found 10 test failures on Blacksmith runners:

Failures

Test	View Logs
/tmp/tmpvvofv34k/summary-fail.json (live):/summary-fail.json (live):	View Logs
/tmp/tmpw6j17yjc/summary-missing.json (live):/summary-missing.json (live):	View Logs
test/agent-pane-background-tasks.test.ts/agent-pane-background-tasks.test	View Logs
test/agent-pane-background-tasks.test.ts/agent-pane-background-tasks.test	View Logs
test/menu-focus.test.ts/menu-focus.test	View Logs
test/menu-focus.test.ts/menu-focus.test	View Logs
test/secret-card.test.ts/secret-card.test	View Logs
test/secret-card.test.ts/secret-card.test	View Logs
test/ui-a11y.test.ts/ui-a11y.test	View Logs
test/ui-a11y.test.ts/ui-a11y.test	View Logs

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

[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]

Shipped in:

• cmux NEXT nightly 1.0.0-nightly.3953592002001 (f3ecea5132)
• cmux NEXT nightly 1.0.0-nightly.3953614980201 (ff8d849c4c)