Startup reads the VM records main keeps on disk before hydrating the session. A selected Orca Cloud workspace whose record is suspended gets sleep intent and the view-only hold before any pane mounts, so its panes wait for Wake instead of reattaching (which woke the sandbox). Running records are untouched. Main now notifies phase 'suspending' when it passes its last reopen check; the view then offers Wake to continue instead of a Keep awake that can no longer work.
* feat(cloud-skills): one row for a skill both agents have; long Synced list starts collapsed
* fix(cloud-skills): Synced disclosure uses the quiet/statusAction button, not restyled ghost
* feat(cloud): sync Claude Code and Codex skills and slash commands to cloud workspaces
Copies ~/.claude/skills, ~/.claude/commands, ~/.agents/skills, ~/.codex/skills, ~/.codex/prompts and untracked project .claude/.agents items into each Orca Cloud workspace on create and wake, next to MCP sync. Plan-then-put host RPCs (agentConfig.planSkillSync / agentConfig.putSkills) with a per-computer ledger: items made or edited in the sandbox are never touched; dropped items are removed. Likely secrets stay on this computer; 256 KB/file, 512 KB/skill, 8 MB/workspace. Optional cross-agent sharing. Settings > Orca Cloud > Skills & commands.
* fix(cloud-skills): design review: Claude naming, grouped list, copies under originals, off state, relative recency
* fix(cloud-skills): use the shared cloud relative time (synced 2m ago)
* fix(cloud-skills): premerge findings (process-host, async prefs/syncer id, exhaustive switches, no casts, runtime catalog)
* fix(cloud-skills): place Skills & commands after Cloud secrets (design-squad nav order)
* chore: restore integration's copies of files the commit hook reformatted during merges
Hides every desktop entry point (Settings section + search, Home's Routines button, Routine runs switch and rows) behind CLOUD_ROUTINES_UI_ENABLED=false. The control plane, API, MCP tools and CLI are unchanged, so existing routines keep running.
Uses the roster's pictureUrl (stablyai/orca-cloud#740; https only, re-checked in main). Initials
stay underneath and show if the image is absent or fails to load; no referrer is sent.
- Main: Cancel aborts the attempt at any stage and clears progress at once; a newer sign-in cancels
the older one (its loopback listener closes); sign-out cancels a pending sign-in; a timeout says
'Sign-in didn't finish. Try again.' instead of an error toast.
- Setup page: a cancelled/timed-out/replaced sign-in no longer falls back to opening a second
sign-in (the root cause of the stuck 'Signing in…'); the waiting row has Cancel; Back cancels a
sign-in the page started.
- Sign in buttons stay clickable while a sign-in is pending.
The control plane sends a run's triggerDetail as an object ({requestedBy}, {event, target}, {scheduledFor}); the desktop schema wanted a string, so POST /run and run history failed to parse although the run had started, inviting a duplicate run. Accept the object and show one short label.
- no Always allow for lookalikes, pipe-to-shell, sudo or command substitution (main refuses it too);
pipe-to-shell warning; tabs and trailing spaces shown; Always allow tooltip; alertdialog + labelled command box
- agent shown by catalog name; recent requests read Finished / Failed (exit N)
- the 3-waiting cap counts requests still being checked (burst); more invisible ranges refused
- Turn off from the dialog reports a failure; grant-device check documented as best effort
- orcad: only the owner's desktop may attach/settle; devices paired through a grant are listed
by orca-sandbox-grant and refused (forks keep their offer's origin id)
- desktop: per-workspace limits (3 waiting, 60/hour), always-allow runs one at a time and 30/hour,
one banner per burst, refusal floods logged once a minute
- dialog: commands over 20 lines refused; long ones must be scrolled to the end; agent name muted;
Turn off for this workspace; read delay counts from when a prompt reaches the front
- lookalike spaces and combining marks refused; settled events only close their own workspace's prompt
- executor reports soon after the shell exits even if an escaped process holds the pipes; hard stop
- audit log keeps 3 files, long commands stored as prefix + sha256
- rules expire after 30 days and are cleared when a shared workspace is allowed
- orcad: files.statMany (batched stat, up to 2,000 paths) for cheap manifests; older images fall back to files.stat
- main: per-workspace sync sessions (full pass on start/reconnect/wake, files.watch-driven incremental passes),
read-only copies under ~/orca/cloud/<repository>/<workspace>, gitignore-aware via files.listAll, bounded
(.context 250 MB / 50 MB per file; mirror 2 GB / 100 MB per file / 100k files), never wakes a sandbox,
stops when asleep, removes the copy on archive/delete
- window: 'On this computer' section in the cloud status panel, '⋯ → Mirror to this computer…',
removal line in the archive toast and delete dialog, confirm before removing changed copies
Plan & billing now reads '<Team> · Team plan', shows the last 30 days of sandbox
hours as a bar chart, and (owners/admins) the top workspaces and members. Members
see the team chart and their own hours. The overage toggle becomes one 'Monthly
spend cap' field ($0 = off, 'Turn on with a $25 cap'); the rate is a one-line
hint; alert levels are read-only values. The manual refresh icon is gone.
Billing Worker: additive GET /v1/billing/usage (read-only over usage_events,
no migration).
- POST /v1/sandboxes takes an Idempotency-Key (api_idempotency, 24 h, scoped per account and member):
the same key and body return the first create's sandbox instead of starting another; same key with a
different body is a 422. The desktop sends context.idempotencyKey when its caller provides one.
- The desktop sends the workspace name as the create's title, so lists show it, not an untitled row.
- GET /v1/sandboxes/boot-plan?repo= says whether a create would clone or update (environment build);
the driver asks in parallel with the create, so the build sheet names 'Clone repository' or 'Update
repository' once, with a placeholder until it knows, and never relabels mid-run.
Home's team list leaves routine-run VMs out unless the new Routine runs switch is on. The switch is
offered only when the team has routines, counts the running runs it hides, and is remembered per
account. Hidden runs leave the workspace count but keep their sandbox hours in the total. Listed
runs carry a routine mark and 'Routine · <name>'. Control plane: GET /v1/team/workspaces
?routines=include, plus fromRoutine/routineName per row and a routineRuns summary.
Design-squad review of #336: a teammate's heatmap is shown only if they opted in (same
Experimental switch as Pokes, renamed 'Pokes and activity'); otherwise their profile says they
haven't shared it. Your own heatmap always shows. Team rows get 'View profile' first in the menu.
Own-profile hint: 'Hover a day to see your sandbox hours.'
- orcad relays (localCommands.request/status/attach/settle); the desktop that owns the workspace decides
- off by default per workspace; owner only; shared workspaces need the owner's explicit OK
- approval dialog: full command and folder, Allow once / Always allow this command / Deny
- minimal env (PATH, HOME, locale), /bin/sh, no stdin, timeout, 128 KB output cap per stream
- audit log on this computer; Settings > Orca Cloud > Commands on this computer
- run_local_command MCP tool (orca local-command mcp) registered for Claude Code and Codex in cloud VMs
Conductor 0.91 parity. Clicking a teammate opens their profile, which shows their role, a
26-week daily usage heatmap and their recent team workspaces. Teammates see only 0-4 levels,
and amounts are shown only to you. Pokes are opt-in in Settings > Experimental and need both
people to have them on. They are rate-limited to 1 per minute per teammate and 20 per hour, and
delivered in-app by a 30 s poll that runs only while Pokes are on and Orca is in front.
Control plane: migration 0050 (member_usage_hours, member_poke_prefs, member_pokes),
src/team-presence.ts routes. The heatmap reads usage_outbox sandbox time plus dated
token_usd_micros deltas. A member's poke opt-in is dropped when they are removed.