* feat(mobile): declare the orca-mobile-web-shell TS surface Two props and one event: a generation directory the TypeScript store owns, a session id that scopes the private origin, and a load state. No module functions and no reload — a retry is a remount under a new React key, which rebuilds the WebView and reinstalls every fence. The native event body is a flat dictionary, so parseMobileWebShellLoadState rebuilds the union instead of asserting it and answers null for anything it does not recognise. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * feat(mobile): serve a generation from a private origin on iOS A WKWebView behind a custom-scheme handler that answers only from a map built once from the generation's manifest, with the CSP as a response header on the document. The scheme handler reads on a serial background queue and keeps a live-task set that stop() removes from: an asset is up to 10 MiB, and delivering to a stopped task raises an Objective-C exception Swift cannot catch. Origin, request refusal, the manifest map and the policy header hold no WebKit type, so tests/MobileWebShellChecks.swift compiles and runs them with swiftc, no device. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * refactor(mobile): compare the iOS shell's applied props field by field One joined string could not tell a directory ending in the separator from a shorter one with a longer session id. Two fields have no separator to collide on. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * test(mobile): pin that a string schemaVersion is not a manifest The contract declares a number. The Kotlin side read it with optInt, which coerces "1" to 1, so a manifest that widened the field would have been served; this check covers the same shape on both platforms. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * feat(mobile): serve a generation from a private origin on Android A WebView behind shouldInterceptRequest, answering only from the same manifest-built map as iOS, with the CSP as a response header on the document. The origin host label is a slice of the session id's SHA-256, never of the session id: Chromium lowercases an https host and java.net.URI reads null for a label holding '_', which is how the reference 403'd every asset. A main-frame failure is reported from a post() because Chromium commits its own error document after onReceivedError returns. onRenderProcessGone destroys the dead WebView and does not rebuild it, so the retry policy stays in one place. clearCache(true) is never called: it is process-global and would wipe the terminal WebView's cache too. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * chore(mobile): untrack the shell module's gradle build output The previous commit staged 312 files from android/build. mobile/.gitignore anchors /android/ at the mobile root, so a module's own gradle output was never covered. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * refactor(mobile): parse the shell load-state payload with a zod shape The anti-slop gate rejects an `object` parameter and `Reflect.get`. zod reads a shape key straight off the value, so the own-property strip stays. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * feat(mobile): give the web shell one load-state machine per platform A failure is terminal, and a repeat says nothing. Chromium commits its error document after onReceivedError returns and a rule list compiles long after a generation was refused, so both platforms could report over a failure the caller had already acted on. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): stop the Android shell reporting ready over a failed document onPageFinished ran after reportDocumentFailure's post and both emitted `ready` and set the WebView visible again, putting Chromium's error page on screen. A prop change after the renderer died now reports instead of going silent. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): publish the Android shell's served generation atomically The map and the host it is keyed against were two plain fields written on the main thread and read on Chromium's, so an interceptor could see a stale null and 403 a good frame, or a new map against the previous host. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): keep the iOS shell to one terminal load state A rule list that failed to compile after a generation was already refused emitted a second, contradictory reason. The document-failure flag it carried is now the state machine's. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): stop serving the previous generation after a failed prop update Both platforms returned early with the old map still installed and the old page still on screen, so a caller told the shell had failed was looking at a working one from the generation before. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): serve the shell document at "/" and nowhere else /index.html answered the same bytes without the policy header, which rides the document response alone. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * test(mobile): pin the shell's response headers as a pure predicate Which response carries the policy header was decided inside the two request handlers, where no test without a device can reach it. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * test(mobile): pin the shell's path-length edge and its charset casing Both limits were checked only from the rejecting side, so a one-off length and an uppercase charset passed unnoticed. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * chore(mobile): state the Android shell's file-URL settings and what B4 must check The two file-URL settings were left to their defaults, and the settings that only a device can prove named nobody to prove them. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * chore(mobile): drop the shell module's unresolved entry points Nothing imports the module by name, on either side; the TypeScript is reached by path, as the notification-dismissal module's is. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): stop the iOS shell failing a document it cancelled itself stopLoading on a prop update and every navigation the policy delegate refuses reach the failure delegates as errors, so a healthy page reported `failed`, lost its `ready`, and sent the caller to delete a good cached generation. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): answer when the iOS rule list store is missing Optional-chaining past a nil store ran no completion handler, so the view stayed at `loading` for good. The next prop update now reads the same terminal isolation failure a compile failure sets. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): refuse a manifest whose schemaVersion is true or 1.0 on iOS NSNumber bridges both to 1, so `as? Int` accepted a manifest Kotlin rejects. Verified against JSONSerialization: objCType is c for true, d for 1.0, q for 1. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): drop an Android document failure the next load did not have The report is deferred past Chromium's error document, so a prop update could land between the decision and the report and fail the generation that had just replaced the one that actually failed. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * test(mobile): assert each blocked global's descriptor whole contains("writable:false") passed on a WebSocket descriptor that had lost it, because the serviceWorker copy still carried one. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * test(mobile): pin the Android shell's navigation and refusal decisions Both lived inside the WebViewClient, which no suite compiles, so dropping the navigation guard or answering a refusal with 200 changed nothing anyone could see. Claude-Session: https://claude.ai/code/session_01JNnE9qzUZMMnqpZWCqM3nb * fix(mobile): name the domain a policy-cancelled frame load is reported under WKErrorDomain has no frame-load codes: WKErrorCode stops at the app-bound domain errors, and 102 belongs to the legacy WebKitErrorDomain. The iOS SDK exports no symbol for it, so the assert that pinned one is gone. 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