[teamleaderleo]

Pilot blocker 1b, app half. When you hover a Cloud machine's sidebar header, the tooltip now shows what the machine costs and the account's machine cap. cloud machine-status also returns the cost to agents.

What a person sees

Hovering a Cloud machine's header in the sidebar shows this under its status:

Connected
$0.26/h · Running 2h 14m · About $0.59 so far
2 of 3 machines running

• A paused machine shows only $0.26/h when running, with no running time or estimate.
• If the plan has no cap (limits.maxActiveVms is null), the cap line reads N running, no machine cap. If the cap is 0, it reads Machines are closed on this plan.
• The running time and estimate are recalculated every time you hover, so they never go stale.
• VoiceOver reads the same lines (accessibility help).

Why the header tooltip: it is the existing place a person hovers over one machine. The rest of the app has no machine hover card. Machine settings show only a name and a connected flag. Cloud Machine Status is a 12 s toast that shows only one line. The header is also where a person looks before resuming or starting another machine.

Where the numbers come from

Everything comes from the server; the app has no rate table of its own:

• The rate and the start time are new fields on each machine from /api/vm: hourlyRateUsd and runningSince. The server PR is cmux Cloud: /api/vm machines carry their hourly rate and when they started running #23682, which takes the rate from the shared rate table in web/services/billing/chiefPricing.ts. Until cmux Cloud: /api/vm machines carry their hourly rate and when they started running #23682 deploys, these fields are absent and no cost line shows.
• The cap comes from the limits object of /api/vm: maxActiveVms and activeVmCount. The server already sends it; the app now decodes it and keeps it on MachineRegistry.cloudLimits. The cap is never inferred from the machine list (fix(cloud): a zero machine cap is not a fraction #16012).
• If a value is malformed, only the cost line is dropped; the machine list still loads.

Agents

cmux cloud machine-status --target machine:vm-… now also returns hourly_rate_usd, running_since (ISO 8601) and estimated_cost_usd, and its text gets the same cost line. cloud list-machines (#23570, not merged yet) should return limits {max_active_vms, active_count}. That follow-up depends on #23570, so it's not in this PR and I didn't touch the files #23570 changes.

Proof

CloudMachineCostTests (CmuxNextAppTests) covers:

• rate and estimate formatting ($0.26, 2h 14m, 3d 4h, under a minute shows as 1m, under one cent);
• the cap for null, 3 and 0, and that it comes from limits, not from counting the list;
• a paused machine shows no running cost, in both the tooltip and the CLI fields;
• decoding, including malformed values;
• the header tooltip, and that an SSH header ignores cost;
• the machine-status fields.

Fleet steps (package-test-lane.sh suite, Xcode 26.6): red c8bca577a9c2e9ee68c0c26a at the test-only commit 462640f (CloudMachine has no member hourlyRateUsd, no MachineCostText); green 2544f59e2c85ba6cca8886db at eac5d29, the last commit with PR changes. That run covered CloudMachineCostTests, SidebarIconMeaningTests, CloudMachineCapabilityTests, CloudCLITargetTests and HostGroupHeaderStatusTests, and all passed. The head 9e3c63f only adds a base merge of feat-cmux-next on top of it. cmux-ci compile-check --app passed at that head (fleet job 9fed05893ce835fc4b7c0eed).

Note on the estimate: the rate is the provider's list price for the machine's size. Plans are flat today, so the line is an estimate of compute used, not a bill.

Strings are localized in all 21 sidebar-catalog locales. check-l10n.sh reports 0 errors and prepush-gate.py is clean.

Capture: native UI, but the tooltip only shows on hover over a live Cloud machine with server cost data, and no tour scene covers that, so I added the media-tour label.

🤖 Generated with Claude Code
Agent session: cosmic-gray

[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.

[teamleaderleo]

Owned by cloud-team-ready on cosmic-gray · session

[github-actions]

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

Review 9e3c63f: 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=pr23691 -f ref=cloud-machine-cost-line -f platform=macos -f backend_mode=local

Passes: CI passes on 9e3c63f547.

CI passes on 9e3c63f547 (run 39705346622 attempt 1).

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: cmux-next checks (god files, concurrency, crash safety, l10n); no finished base run on db5d0b0 to compare; do not merge until green; see lane-rules

Tour of 9e3c63f547 (run): no media. The tour job failed before the tour ran (build or setup); the run log has the output.

[blacksmith-sh]

Found 11 test failures on Blacksmith runners:

Failures

Test	View Logs
/tmp/tmp0alqqr9m/summary-missing.json (live):/summary-missing.json (live):	View Logs
/tmp/tmpkae594jj/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
convention violations found (ERROR lines above)./ convention violations found (ERROR lines above).	View Logs
nightly cmux-tui resolve/nightly cmux-tui resolve	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

...and 1 more test failures. View all on Blacksmith

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

[teamleaderleo]

Bead cx-hxr9t created for this PR by the cmux coordinator.

[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.

[teamleaderleo]

cmux-next guards / nightly-guards / workflow-guard-tests / release-notary is red on 14 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.

[teamleaderleo]

cmux-next guards / cmux-next localization debt ratchet 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.

[teamleaderleo]

cmux-next crash-class rules / cmux-next crash-class rules is red on 11 of 11 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]

Review (subagent, correctness first): SHIP at 9e3c63f

The review covered the PR diff against the base it merged (db5d0b0..9e3c63f) and found nothing blocking. cmux-ci compile-check --app passed at this head (fleet job 9fed05893ce835fc4b7c0eed). The cost suites passed in fleet step 2544f59e2c85ba6cca8886db at eac5d29, the last commit with PR changes. Remaining reds, none of them from this PR:

• cmux-next crash-class rules: CEFFileDragRelay.swift:162 needs a selector probe. It is red on base too: https://github.com/manaflow-ai/cmux/actions/runs/39703291655/job/119577318988
• cmux-next checks, "Checks job structure" (test_every_submodule_free_checkout_on_an_owned_runner_resets_submodules), is red on base too: https://github.com/manaflow-ai/cmux/actions/runs/39708091959/job/119592751426
• release-notary and Guard status are red on base too: https://github.com/manaflow-ai/cmux/actions/runs/39703950507/job/119579454354
• cmux-next generated files: plans/cmux-next/action-surfaces.json is stale. The regenerated file differs from the base's committed copy by the same 10 lines, and this PR does not touch actions.
• cmux-next swift test is still queued because the glaeda-gui-std pool is backed up for every PR.

[github-actions]

Shipped in:

• cmux NEXT nightly 1.0.0-nightly.3973137732601 (15186f150b)
• cmux NEXT nightly 1.0.0-nightly.3973389583201 (48a69e9f66)