feat(mobile): full-screen simulator viewer with on-demand controls

[juliusmarminge]

The mobile device viewer spent a lot of the screen on chrome: a stack header with close, Home and the menu, safe-area padding, and the app's status bar. That left the simulator stream small and letterboxed.

The stream now takes the whole screen at the device's own aspect ratio. The header, status bar, home indicator and Android navigation bar are hidden. Close, the device name, Home and the … menu (device switching, reload, back, app switcher, rotate, tool versions, shut down) live in a glass overlay:

• It shows while the stream connects, then hides once touch input connects.
• Tapping the dimmed area hides it again.
• To bring it back, tap the small handle at the top centre, or shake the phone. A shake is two strong jolts within 0.6s, with a light haptic tap.
• On Android, the back button first reveals the controls rather than leaving.

Shake detection uses expo-sensors (new native module, so a new binary rather than an OTA update). The detector is a pure function with tests.

Before / after

Captured on an iOS simulator streaming another simulator from an isolated backend.

Before	After (controls hidden)	After (controls shown)	After (menu)

The "after" captures are from a second iPhone 17 Pro simulator, whose own status bar and notch sit at the top of the stream. The connecting state:

Flow recording (open viewer, show/hide controls, menu, close):

https://gh-file-drop-api-prod-mi5fy3sowv63ufte.pinglabs.workers.dev/f/f0067dc2a27692df/flow.mp4

Verification

• vp test run src/features/devices/shakeDetector.test.ts, mobile typecheck, and lint on changed files.
• Driven end to end on an iOS simulator. The simulator has no accelerometer, so shake itself is not verified there; its "Shake" opens the Expo dev menu in dev builds. EAS preview builds for iOS and Android went out for on-device testing.

Claude Opus 5.5 (1M context) in Claude Code.

🤖 Generated with Claude Code

---

[macroscopeapp]

Approvability

Verdict: Not approved

Macroscope's review found this PR not approvable — The PR changes the existing device-preview route into a full-screen experience with hidden system UI, auto-hidden controls, accelerometer-triggered toggling, and new native sensor configuration. These are default production behavior and interaction changes rather than an isolated opt-in addition.

^{You can add or adjust custom eligibility rules. Learn more.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	4.9 KiB	4.9 KiB	−40 B (−0.8%)	6.8 KiB	✅
Codex	Thread snapshot wire	3.7 KiB	3.7 KiB	0 B (0.0%)	4.9 KiB	✅
Codex	Live turn WebSocket wire	1.2 KiB	1.1 KiB	−40 B (−3.3%)	2.0 KiB	✅
Codex	Live turn WebSocket decoded	20.4 KiB	20.4 KiB	−41 B (−0.2%)	29.3 KiB	✅
Codex	Live turn messages	2	1	−1 (−50.0%)	8	✅
Claude	Total thread wire	4.9 KiB	4.9 KiB	−41 B (−0.8%)	6.8 KiB	✅
Claude	Thread snapshot wire	3.7 KiB	3.7 KiB	0 B (0.0%)	4.9 KiB	✅
Claude	Live turn WebSocket wire	1.2 KiB	1.2 KiB	−41 B (−3.4%)	2.0 KiB	✅
Claude	Live turn WebSocket decoded	20.8 KiB	20.7 KiB	−41 B (−0.2%)	29.3 KiB	✅
Claude	Live turn messages	2	1	−1 (−50.0%)	8	✅

Baseline: 1302cca · PR result: 3bbec85 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 106.1 KiB
• Claude decoded thread snapshot: 106.4 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.