- Command-finished events carry the pane's leaf; a run learns its pane while its tab has one,
ignores other panes, and Stop interrupts the run's own pane.
- Workspaces whose setup ran in the environment build (hooks report environmentProvisioned) start
the autoRunAfterSetup action the first time they are opened, once per workspace.
* fix(w8): DS-41 phone polish: grey backdrop behind white groups in light, light amber at 3:1+, access rows read '{device}, last used {date}' with a 'Removes this phone's access' hint
* fix(w8): DS-41 review: white text on light amber badges (onStatusAmber, 5:1), grey backdrop only behind grouped lists (card screens keep white), 'connected {date}' label when a phone was never used
* feat(w8): DS-28 new cloud workspace: pick the repository from this phone's recents and the team's projects ('Other repository…' for any other), and say 'Sleeps after N minutes idle' (new screen, empty Home)
* copy(w8): DS-28 with the picker, the note says 'Pick none for a blank project'
* fix(w8): DS-28 review: explicit 'No repository (blank project)' row, the field stays while typing an unlisted repo, choosing 'Other repository…' again keeps the text
* fix(w8): DS-28 nit: typing in the field selects 'Other repository…'
- nonconcurrent now matches Conductor's run_mode: starting it stops that action in the project's
other workspaces; the project's other actions keep running (W9 finding).
- A VM launcher's ORCA_PORT=3000 is the first workspace's block (persisted, stable across wake), so
its agents, terminals, setup and run actions all agree; later worktrees on the VM get registry
blocks that never overlap it. The daemon no longer hands the host's block to panes without one,
and setup/archive hooks resolve the workspace block before they run.
- local.run-actions: autoRunAfterSetup after a good setup and not after a failed one, Run button,
one trust prompt, Stop -> Stopped, menu Restart replaces the tab, Cmd+Opt+Enter start/restart
- run.nonconcurrent: echo-loop started in B stops A's
- run.ports: ORCA_PORT/ORCA_PORT_COUNT/CONDUCTOR_PORT per terminal, distinct blocks, stable across
an app restart, served locally; cloud half: the action in the VM reachable here via auto-forward
- run.cloud-resume: action back after Sleep/wake and after Restart sandbox
DS-34:
- One billing store (billing-summary-store) replaces the summary, the hook's local state and the
Run-on picker cache; it keeps the last good result through a failed read and still drops answers
from before a team or account switch (DS-17).
- Skeleton while Plan & billing loads.
- '{{count}} sandbox hours left' and '{{count}} sandbox hours over the included amount this month'
as i18n plurals; no bare 'hours'; the unreachable 100% branch is gone.
- Usage warning driven by usageWarningPercent ('Your team has used {{percent}}% of this month's
sandbox hours.'), in Settings, the plans page and the account menu.
- capTooHigh states the maximum ('The spend cap can be at most {{max}}.'), from offers.maxSpendCapCents.
- Plans dialog focuses its heading on open, so Tab reaches the interval toggle before the cards.
- Payment notice uses InlineError; the build card's overage buttons land on the overage section.
- Account menu names the team once and adds the hours-left line.
DS-44: warning tone at 90%+, '$' as an input prefix, neutral paid check, 24px dialog close hit
area, no <p> inside the CheckoutPanel <dl>.
- Agent and worktree come from the terminal orcad verifies from launch
evidence (verifiedCallerTerminalHandle, shared with turn-sender); the
terminalHandle param is gone and a worktree hint is kept only when the
host knows it.
- secrets.* are accepted, and their events sent, only on cloud orcad
(ORCA_IS_CLOUD=1 + ORCA_SANDBOX_ID); already-set never reads orcad's env.
- touchRequest emits secrets.requestUpdated; natural expiry emits
secrets.requestSettled {status: expired}.
- orca-env skips the control plane's reserved names and prefixes, with a
test that reads them from secrets.ts.
- reason drops bidi controls and turns C0/C1 controls into spaces.
Customers keep flat sandbox hours. Each stopped session's activeCpuDurationMs
(from Vercel's stop response or GET session) is stored in a new session_cpu
table (migration 0042), idempotent per session. Recording is best-effort and
never blocks suspend, stop or usage reporting; a 5-minute cron sweep retries
recent sessions still missing CPU. Operator-only GET /v1/internal/fair-use
reports per-account 30-day utilisation sum(active_cpu_ms)/sum(wall_ms*vcpus)
with fairUseFlag above 60%. Nothing reaches usage_outbox, billing or clients.
- Orca account pane: signed-out and reconnect-required lines name Orca Cloud.
- Every unauthorized failure (lifecycle toast, cloud workspace errors, share links,
cloud secrets) now says "Sign in to Orca again to continue." with a "Sign in" action.
- Team page busy row keeps full contrast: spinner + aria-busy, no opacity dimming.
- destructive-chip text uses a new --destructive-text token (light: 85% mix with
foreground) so it clears AA on its tint.
- TeamDialogHost quietRefresh depends on the stable refresh callbacks.
#216 let a launcher-set ORCA_PORT (the VM launcher sets 3000) override the per-workspace block, so
two worktrees on one cloud VM would both get 3000. Each workspace always gets its own registry
block again; a regression test covers a host env with ORCA_PORT=3000.
Live on dev the first capture sealed zero sessions: agentSession.subscribeStatus needs a streaming
transport and the unix socket refuses it, and worktree.ps lists only agents live right now. The
collector now reads orcad's own session records (agent-session-journal.db, read-only): provider,
name and last turn, open tabs first; live worktree.ps rows override status; history still comes
from agentSession.history, with the stored last turn as fallback.
* fix(w3): port row keeps only Open in browser, Share… and More inline; Copy details gets its own icon (DS-29 follow-up)
* chore(w3): drop the unreferenced inline copy-button label from every locale
Found live on an Orca Cloud rollover (m4air): the persisted-run reader referenced a const declared
below the module-level read, so every restart started with no runs and nothing relaunched. The
relaunch also read orca.yaml before the replacement host served RPCs; it now retries with a bound.
orcad keeps agent-initiated secret requests in memory with the contract's
limits (3 pending, 10/hour, 15 min TTL, touch capped at 60 min, dedupe per
name), derives the agent from the calling terminal, and announces
secrets.requested on runtime.clientEvents (replayed on attach; phones get
status only). The CLI waits on it and prints the D4 reload hint; the VM
ships orca-env and makes zsh read the session env hook.
Neil
Neil