* refactor(mobile): export the mobile web manifest read schema
The generation store re-parses the manifest it cached, and it must read it back
with the same loose reader the fetch accepted it under: parsing strictly after
accepting loosely would turn a host's added field into a forced redownload on
every launch.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* feat(mobile): add the per-host mobile web generation store
Turns a verified bundle into an atomically activated, host-scoped generation
directory under the OS cache, and reads it back. No RPC, no UI, no flag: the
native view is later handed the directory read-only and never writes to it.
The single directory under `generations/` is the activation, so there is no
activation file to edit: a commit deletes every other generation before the
rename, an interrupted one leaves zero generations for the runbook's redownload
rule, and two directories or an unreadable manifest drop the host tree instead
of guessing. `tmp/` is never an activation candidate and every one of them goes
at launch. `hosts.json` carries recency only, so losing it costs eviction order
rather than a generation.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): never evict the host a commit just activated
`now()` is a wall clock. With four hosts cached, one backward jump made the
fifth commit's own entry the oldest, so it evicted the host it had just
activated and handed back an ActiveGeneration whose directory was gone.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* Revert "fix(mobile): never evict the host a commit just activated"
This reverts commit a082ac1777. That commit carried all six round-1 fixes under
a subject naming only one of them; the six land again below, one per commit.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* refactor(mobile): drop the generation store's unused eviction entry point
`evictHostsBeyond` had no caller: commit enforces the four-host ceiling itself,
and a launch-time sweep for a shrunk limit can be added when something shrinks
it. The two `createDirectory` calls went with it, since the port already creates
intermediates, plus a line on what the Android rename fallback leaves behind.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): never evict the host a commit just activated
`now()` is a wall clock. With four hosts cached, one backward jump made the
fifth commit's own entry the oldest, so it evicted the host it had just
activated and handed back an ActiveGeneration whose directory was gone.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): fold case when refusing an asset named like the manifest
APFS and NTFS are case-insensitive by default, so `Manifest.JSON` landed on the
store's own `manifest.json` and the activation read back as the asset's bytes.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): keep a completed activation when the recency index cannot be written
`hosts.json` is written after the rename, so a disk that filled between the two
turned a generation already on disk into a thrown commit. The index carries
recency, not truth, and the next activation rewrites it whole.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): refuse to commit a staged handle whose tree is gone
Commit deleted every other generation before it looked at the staged tree, so
committing an aborted or swept handle destroyed the live generation and only
then threw. The check moves ahead of the first delete.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* test(mobile): close four surviving generation-store mutants
Sweeping only the first host's tmp, staging over residue, dropping the serial
queue, and dropping the stale-index pruning all passed the suite. The stage
race needed two differing asset lists under one build id to be visible at all:
with identical ones an interleaved pair ends on the same bytes as a serial one.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): activate only a build-id entry that holds a manifest
An entry under `generations/` matching the staged build id was taken as the
activation on its name alone, so an empty directory of that name — what a
crash between the rename and the post-rename check leaves on Android under
API 26 — or a plain file made the commit drop the verified staged tree and
return a generation that cannot be read back.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): never delete a cache because a manifest read failed
The adapter mapped every `file.text()` throw to null and the reader treated
null as corruption, so one iOS data-protection or I/O blip deleted the only
verified generation a host had. Missing stays null and still drops the tree;
a failed read now throws, and the reader returns no activation without
touching disk, leaving the caller to redownload.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* refactor(mobile): drop the generation store's unreachable build-id guard
`MobileWebBundleManifestReadSchema` already pins `buildId` to the sha256
pattern, so no manifest reaching the store can fail the second check.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* test(mobile): pin that eviction ignores a non-host directory
Dropping the host-key filter in `listHostDirectories` passed the whole suite;
the ceiling would then count and evict anything else under the OS cache
directory.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): keep a mid-download host out of the cache ceiling
A host holding only a staging tree was counted against the four-host limit
and, having no index entry, sorted first for eviction, so four cached hosts
plus one download meant the next activation deleted the tree that download
was about to commit. The ceiling now counts hosts with a generation; sweeping
still walks every host directory.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): record recency when a commit finds the build already active
The same-build early return skipped the index write, so a host that
redownloaded the bundle it already had stayed the least recently activated
and was the first evicted. No eviction pass on that path: the host count is
unchanged.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
* fix(mobile): honour only staged handles the store itself issued
`StagedGeneration` is structurally typed, so any object of that shape made
`commitGeneration` rename over, and `abortStagedGeneration` delete, a
directory of the caller's choosing. Handles are tracked in a per-store
`WeakSet` and anything else is refused before a filesystem call.
Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb
Orca Mobile
React Native companion app for Orca. Monitor worktrees, view terminal output, and send commands from your phone.
Local development uses two processes:
- Orca desktop/Electron from the repo root. This hosts the mobile WebSocket RPC server on port
6768. - Expo Metro from
mobile/. This serves the React Native app on port8081.
Unless a command says otherwise, run mobile app commands from the mobile/ directory.
Prerequisites
- Node.js 24+
- pnpm
- Xcode and/or Android Studio tooling for simulator or device builds
- Expo Go on your phone, or a development client build when native modules are needed
- Phone and desktop on the same LAN when testing a physical phone
Start Desktop Orca
From the repository root:
pnpm install
pnpm dev
Confirm the mobile RPC server is listening:
lsof -nP -iTCP:6768 -sTCP:LISTEN
Restart pnpm dev after changing Electron main-process code. Metro hot reload only applies to the mobile JavaScript bundle.
Start The Mobile App
cd mobile
pnpm install
pnpm start
Scan the Expo QR code with your phone's camera on iOS, or Expo Go on Android.
For a native dev-client build:
pnpm exec expo run:android
pnpm exec expo run:ios
pnpm start --dev-client
Pair With Desktop Orca
- Open Orca desktop.
- Go to Settings > Mobile.
- Scan the pairing QR code from the mobile app.
- Confirm the mobile host endpoint is
ws://<desktop-ip>:6768.
For the Android emulator, use ws://10.0.2.2:6768. For a physical phone, use the desktop LAN IP, for example ws://192.168.0.179:6768.
If the phone has a stale host entry, remove it from the app and pair again.
Development Paths
Android Phone
- Install Expo Go from Google Play
- Run
pnpm start, scan QR with Expo Go - For native modules:
pnpm exec expo run:android - Run with
pnpm start --dev-client
iOS Simulator
- Install Xcode from the App Store
- Run
pnpm start --iosto open in iOS Simulator
Physical Phone Debugging
The phone can be inspected through the connected device tooling:
orca snapshot --json
orca click --element @e3 --json
orca fill --element @e1 --value "ls" --json
orca screenshot --json
Use snapshot first to find the current element refs, then click/fill those refs. After mobile file edits, Metro usually hot reloads automatically, but navigating out of and back into the session screen can be useful because it re-runs terminal.subscribe.
Terminal Streaming Repro Without A Phone
Use this when terminal output does not render on device and you need to split server streaming bugs from WebView/UI bugs:
cd mobile
ORCA_MOBILE_WS_URL=ws://127.0.0.1:6768 pnpm exec tsx scripts/test-subscribe.ts <deviceToken> <serverPublicKeyB64>
You can pass a worktree selector as the third argument:
pnpm exec tsx scripts/test-subscribe.ts <deviceToken> <serverPublicKeyB64> "id:<worktreeId>"
pnpm exec tsx scripts/test-subscribe.ts <deviceToken> <serverPublicKeyB64> "path:/absolute/worktree/path"
pnpm exec tsx scripts/test-subscribe.ts <deviceToken> <serverPublicKeyB64> "name:my-worktree"
The expected result includes:
streamSawMarker: true
readSawMarker: true
If this repro fails, debug the desktop runtime/PTY path before the mobile WebView. If it passes but the phone is blank, debug the session screen or TerminalWebView readiness/queueing path.
Terminal Color Repro Without A Phone
Use this when terminal colors disappear after switching tabs. Open a Claude Code terminal and at least one other terminal in the target worktree, then run:
cd mobile
ORCA_MOBILE_WS_URL=ws://127.0.0.1:6768 pnpm exec tsx scripts/repro-terminal-colors.ts \
<deviceToken> <serverPublicKeyB64> "id:<worktreeId>"
The script captures terminal.subscribe snapshots in an A → B → A sequence and writes raw snapshots to mobile/terminal-color-repro/. If the two A snapshots have different sgrColor counts, the desktop snapshot changed during the switch. If they match, the ANSI color data is still present and the bug is in mobile replay/rendering.
Validation
Run these checks before committing mobile terminal changes:
cd mobile
pnpm exec tsc --noEmit
pnpm run check:tests-typecheck
pnpm lint
cd ..
pnpm typecheck:node
tsc --noEmit reads tsconfig.json, which excludes test files so Metro never bundles them.
tsconfig.test.json puts them back, and pnpm run typecheck:tests shows their errors in full.
check:tests-typecheck is the gate over it: a ratchet against tests-typecheck-baseline.txt, the
127 test files that do not typecheck yet. It fails when a file that checks today stops checking,
and when a baseline entry starts checking (prune it with
node scripts/check-tests-typecheck-ratchet.mjs --prune). The list may only shrink.
The same gate censuses the program first: every *.test.ts(x) on disk must be in it, or named in
the script's TESTS_OUTSIDE_PROGRAM with a reason. Without that, a test excluded from
tsconfig.test.json — or a Foo.test.tsx shadowed by a Foo.test.ts beside it, which a wildcard
include drops for the higher-priority extension — would leave the ratchet silently.
Protocol Version Compatibility
Mobile and desktop talk over a versioned protocol. Because mobile updates lag desktop by 24-48h via the App Store, both sides exchange version numbers on status.get so a genuinely incompatible combo can hard-block instead of silently misbehaving.
Constants live in two files (Metro can't resolve outside mobile/):
src/shared/protocol-version.ts—DESKTOP_PROTOCOL_VERSION,MIN_COMPATIBLE_MOBILE_VERSIONmobile/src/transport/protocol-version.ts—MOBILE_PROTOCOL_VERSION,MIN_COMPATIBLE_DESKTOP_VERSION
Today all four are set so evaluateCompat always returns { kind: 'ok' } — nothing blocks. The wire format is in place to flip a switch when needed.
When to bump
Bump DESKTOP_PROTOCOL_VERSION (and the mobile mirror MOBILE_PROTOCOL_VERSION when relevant) for breaking changes:
- Removed RPC method or required parameter that mobile uses
- Changed meaning (units, nullability) of an existing field mobile reads
- Changed encryption, framing, or auth handshake
Do not bump for additive changes:
- New RPC methods
- New optional fields on existing methods
- New event types in
terminal.subscribe
Set MIN_COMPATIBLE_MOBILE_VERSION (kill-switch) when desktop ships a change that requires a minimum mobile version to function safely. Same for MIN_COMPATIBLE_DESKTOP_VERSION from the mobile side.
When a verdict is blocked, mobile/src/components/ProtocolBlockScreen.tsx renders a screen pointing the user at either the App Store (mobile too old) or GitHub Releases (desktop too old).
To exercise the block screen locally: set MIN_COMPATIBLE_DESKTOP_VERSION = 999 in mobile/src/transport/protocol-version.ts, rebuild, pair to any desktop. Revert before merging.
Mock Server
Develop the mobile app without a running Orca desktop instance:
pnpm mock-server # starts mock WebSocket server on port 6768
Connect from the app using endpoint ws://localhost:6768 and token mock-device-token.
Environment variables
MOCK_NATIVE_CHAT=1— serve the native-chat scenario (one live agent tab, empty transcript, image upload) instead of the default terminal fixtures.MOCK_CHAT_AGENT=omp— withMOCK_NATIVE_CHAT=1, present an OMP tab and four decoded transcript messages, including a tool call and result, instead of the default Claude scenario. It deliberately omitstranscriptPathto exercise legacy-hook readability discovery; current OMP hooks may report a path.MOCK_SERVER_KEY_FILE— persist the server keypair across restarts so a paired device keeps its public-key pin. A missing or invalid file is re-keyed with a warning, which forces a re-pair.
Scenario control files
Read on every request, so behaviour can be flipped mid-session without a restart (a restart would re-key E2EE and force a re-pair). Write the mode into the file, or delete it for the default.
MOCK_SEND_MODE_FILE(defaultorca-mock-send-modein the system temporary directory) —accept(default) accepts the send,errorfails it withmobile_input_floor_unavailable, anything else reports the send as rejected.MOCK_TERMINAL_LIST_MODE_FILE(defaultorca-mock-terminal-list-modein the system temporary directory) —omitreturns an empty terminal list,otherreturns a list that omits the chat handle, anything else lists it.MOCK_TERMINAL_STREAM_MODE_FILE(defaultorca-mock-terminal-stream-modein the system temporary directory) —deadanswers a subscribe withsubscribedthenend(a gone PTY), which is what exercises the rearm bound and terminal prune; anything else streams normally.
Connecting to Real Orca
- Start Orca desktop with WebSocket transport enabled
- In Orca, go to Settings > Mobile and scan the QR code with this app
- The QR encodes the connection endpoint, device token, and TLS fingerprint
Project Structure
mobile/
├── app/ # Expo Router screens (file-based routing)
│ ├── _layout.tsx # Root layout with navigation stack
│ ├── index.tsx # Home screen — paired hosts list
│ └── pair-scan.tsx # QR code scanning screen
├── src/
│ ├── terminal/ # Terminal WebView and xterm bridge
│ └── transport/ # WebSocket RPC client
├── scripts/
│ ├── test-subscribe.ts # Desktop streaming repro without a phone
│ └── mock-server.ts # Standalone mock WebSocket server
└── assets/ # App icons and splash screen