testsAndMisc/focus_owner
Krzysztof kuhy Rudnicki 1b7958143d Prove DPM app-hiding survives a reboot; build the enforcement decision layer
The whole Device Owner design assumed setApplicationHidden persists across a
reboot, and that had never been tested -- while both shell primitives were
measured to be cleared by one. Settling it looked like it needed a factory
reset. It does not: a managed profile is removable and takes its profile-owner
grant with it, so the DPM API can be exercised and then erased.

  pm create-user --profileOf 0 --managed   -> user 10
  dpm set-profile-owner --user 10 ...
  ... hide, reboot, re-check ...
  pm remove-user 10

hidden=true before the reboot, hidden=true after, agreed by dumpsys and by the
DPM API itself. Afterwards: Device Owner Type -1, no restrictions, and Chrome
in user 0 untouched.

Scope limit, stated because it is easy to overclaim: this proves the API's
hidden state persists under PROFILE OWNER inside a work profile. It does not
prove a DEVICE OWNER can hide a given app in user 0, which is what the design
actually needs -- DO there additionally cannot touch the launcher, the default
dialer/SMS/IME or packages the system deems critical, and the call can return
false silently. That must be tested in the first minute after provisioning,
while clearDeviceOwnerApp() is still a cheap exit.

On top of that, EnforcementDecision turns policy plus conditions into a
hide/show set. It is a pure function, so the whole decision layer is tested
without a device, a clock or a location provider.

packagesToShow is deliberately the complete allowed set rather than a delta.
Failing to hide something is a mild disappointment; failing to UNHIDE strands
the user without a dialer or a bank at 09:00 because the phone rebooted during
curfew. Applying the full set on every evaluation makes a missed run
self-repairing, which matters because BOOT_COMPLETED does not fire until first
unlock -- observed directly, the phone sat at RUNNING_LOCKED for minutes after
a reboot with no network at all.

The workout exception is expressed as packages rather than domains. The rooted
system unblocked youtube.com in a hosts file; hiding an app cannot target a
domain, so the package set is derived from the configured domains and config.sh
stays the source of truth for intent.

Signal sources were checked on the device before being built: StrongLifts is
granted health.WRITE_EXERCISE and Health Connect is installed, so that path is
viable; com.kuhy.workout_app exposes only a MainActivity and Flutter
boilerplate; org.runnerup is not installed here at all. Only the two that can
actually produce a signal will be wired.

HideTestReceiver is temporary and must be deleted before any provisioned build.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AwepyL5Lxs5QKbKkMpxxxL
2026-08-09 21:36:26 +02:00
..
android Prove DPM app-hiding survives a reboot; build the enforcement decision layer 2026-08-09 21:36:26 +02:00
assets Serve the blocklist over DoT and hand the policy to the app as an asset 2026-08-05 09:36:11 +02:00
lib Prove DPM app-hiding survives a reboot; build the enforcement decision layer 2026-08-09 21:36:26 +02:00
test Prove DPM app-hiding survives a reboot; build the enforcement decision layer 2026-08-09 21:36:26 +02:00
.gitignore Make focus policy and blocklist usable without root; fix workout CDN unblock 2026-08-05 09:15:02 +02:00
.metadata Make focus policy and blocklist usable without root; fix workout CDN unblock 2026-08-05 09:15:02 +02:00
analysis_options.yaml Make focus policy and blocklist usable without root; fix workout CDN unblock 2026-08-05 09:15:02 +02:00
pubspec.lock Make focus policy and blocklist usable without root; fix workout CDN unblock 2026-08-05 09:15:02 +02:00
pubspec.yaml Exempt the generated policy asset from prettier 2026-08-05 09:40:44 +02:00
README.md Make focus policy and blocklist usable without root; fix workout CDN unblock 2026-08-05 09:15:02 +02:00

focus_owner

Device Owner companion app for unrooted focus mode. Currently inert: it can be provisioned as device owner, but it applies no restrictions.

Why the escape hatch ships first

dpm remove-active-admin does not work on a device owner. The only exit that avoids a factory reset is DevicePolicyManager.clearDeviceOwnerApp(), callable solely by this package on itself. An app that locks the device down before proving it can let go leaves a soft-brick recoverable only through fastboot — painful on a locked bootloader, and impossible to reach if DISALLOW_FACTORY_RESET is ever set.

So: DevicePolicyBridge.releaseDeviceOwner() is implemented, unit-tested, and reachable from the app's main screen with no PC attached. Enforcement is added only after that path has been exercised on a provisioned device.

Provisioning (NOT done yet — requires a factory reset)

dpm set-device-owner refuses to run while any account exists on the device. The Pixel 6a has 6, so provisioning means a wipe. See ~/.claude/projects/-home-kuhy-testsAndMisc/memory/device-owner-not-root-banking-works.md for what does and does not survive that, and for the Google Wallet open risk.

Once wiped, before adding any account:

adb install -r build/app/outputs/flutter-apk/app-debug.apk
adb shell dpm set-device-owner com.kuhy.focus_owner/.FocusDeviceAdminReceiver

Then immediately verify tap-to-pay and banking apps, while the release button is still reachable.

Build and verify

cd ~/testsAndMisc/focus_owner
flutter analyze && flutter test
flutter build apk --debug
adb -s 23181JEGR08034 install -r build/app/outputs/flutter-apk/app-debug.apk
adb -s 23181JEGR08034 shell am start -n com.kuhy.focus_owner/.MainActivity

The status screen reads live state over the platform channel; "Android SDK 36" confirms the Kotlin bridge is working.

Still to do

  • Consume the policy JSON from python_pkg/focus_policy (export.py).
  • Add enforcement (setApplicationHidden, addUserRestriction, setAlwaysOnVpnPackage) — only after the release path is verified on-device.
  • Verify setApplicationHidden survives a reboot. pm suspend and pm disable-user both do NOT; do not assume the DPM equivalent behaves the same until measured.