advance_root_detection 0.0.1

Comprehensive Flutter RASP plugin for Android & iOS: detects rooted/jailbroken devices, Frida/Xposed hooking, emulators, debuggers, and app tampering with a hardened NDK-level detection layer.

advance_root_detection

Comprehensive Flutter RASP (Runtime Application Self-Protection) plugin for Android and iOS.

Detects rooted/jailbroken devices, hooking frameworks (Frida, Xposed, Substrate), emulators/simulators, debuggers, and app tampering — with a hardened NDK-level C++ detection layer on Android that significantly raises the bar for bypass attempts.

Disclaimer: This plugin is intended for legitimate app hardening only. Default behavior is to warn or degrade gracefully — never to brick the device or corrupt user data. No client-side RASP is unbypassable against determined attackers. Use in combination with server-side attestation (Apple DeviceCheck / App Attest, server-side certificate pinning).

---

Features

• Root detection (Android): su binaries, Magisk/Zygisk, test-keys, dangerous props, root manager packages
• Jailbreak detection (iOS): classic + modern (Dopamine, palera1n, rootless), URL schemes, sandbox escape
• Hooking detection: Frida (port + process + memory maps), Xposed/LSPosed, Cydia Substrate, libhooker
• Emulator/Simulator detection: Build fingerprint, QEMU props, telephony, sensor count, emulator files
• Debugger detection: Debug.isDebuggerConnected, FLAG_DEBUGGABLE, TracerPid, sysctl P_TRACED
• App integrity: APK/IPA signing cert SHA-256, installer source, bundle ID, package name
• Environment signals: screen capture, accessibility services, VPN, developer mode/ADB
• Android NDK hardening layer: native ptrace, raw-syscall /proc/self/maps scan, inline-hook detection, JNIEnv integrity, XOR-obfuscated strings, .text segment writability check
• Stream-based continuous monitoring via EventChannel
• Configurable per-app: expected package name, signing cert hashes, allowed install sources, bundle IDs

---

Installation

dependencies:
  advance_root_detection: ^0.0.1

Android

Add to your android/app/build.gradle:

android {
    defaultConfig {
        minSdk 21
    }
}

iOS

Minimum iOS 12.0. No additional setup required.

---

Quick Start

import 'package:advance_root_detection/advance_root_detection.dart';

final shield = AdvanceRootDetection();

// One-shot full check
final ThreatReport report = await shield.performCheck(SecurityConfig(
  android: AndroidConfig(
    packageName: 'com.example.app',
    signingCertHashes: ['A1B2C3...'], // SHA-256 hex, no colons
    allowedInstallers: [AppStore.googlePlay],
  ),
  ios: IOSConfig(
    bundleIds: ['com.example.app'],
    teamId: 'ABC123',
  ),
));

if (report.isPrivilegedAccess) {
  // Root or jailbreak detected
}
if (report.isRuntimeManipulated) {
  // Frida, Xposed, or Substrate detected
}
if (report.hasCriticalThreat) {
  // Block sensitive operation
}

// Stream-based monitoring
shield.threatStream.listen((Threat threat) {
  switch (threat.category) {
    case ThreatCategory.privilegedAccess:
    case ThreatCategory.runtimeManipulation:
    case ThreatCategory.debuggerAttached:
    case ThreatCategory.analysisEnvironment:
    case ThreatCategory.integrityViolation:
    case ThreatCategory.screenCapture:
    case ThreatCategory.untrustedSource:
      print(threat);
  }
});
await shield.startMonitoring(SecurityConfig(
  monitoringInterval: Duration(seconds: 30),
));

// Verify before sensitive operations (payments, biometrics)
final safe = await shield.verifyBeforeSensitiveOp();
if (!safe) {
  // Block the operation
}

// Stop monitoring when done
await shield.stopMonitoring();

---

API Reference

AdvanceRootDetection

Method	Returns	Description
performCheck([SecurityConfig])	Future<ThreatReport>	Full security check, all detectors
startMonitoring([SecurityConfig])	Future<void>	Start background monitoring
stopMonitoring()	Future<void>	Stop background monitoring
verifyBeforeSensitiveOp([SecurityConfig])	Future<bool>	Quick check, true = safe
threatStream	Stream<Threat>	Continuous threat events

ThreatReport

Property	Type	Description
isPrivilegedAccess	bool	Root / jailbreak detected
isRuntimeManipulated	bool	Hooking framework detected
isDebuggerAttached	bool	Debugger present
isAnalysisEnvironment	bool	Emulator / simulator
isIntegrityViolated	bool	Signing or bundle integrity violated
hasCriticalThreat	bool	Any high or critical severity threat
isClean	bool	No threats at all
detectedThreats	List<Threat>	All individual findings

ThreatCategory

Value	Description
privilegedAccess	Root (Android) or jailbreak (iOS)
runtimeManipulation	Frida, Xposed, Substrate, etc.
debuggerAttached	Debugger or anti-debug triggered
analysisEnvironment	Emulator or simulator
integrityViolation	Signing / bundle integrity
screenCapture	Screen recording / sharing
untrustedSource	Unknown or untrusted install source

Severity

info → low → medium → high → critical

---

Detection Capability Matrix

Category	Detection Method	Android (Kotlin)	Android (NDK)	iOS (Swift)
Root / Jailbreak	su binary (8 paths)	✅	✅	—
	BusyBox	✅	—	—
	Magisk artifacts (6 paths)	✅	✅	—
	Test-keys in Build.TAGS	✅	—	—
	Dangerous props (ro.debuggable, ro.secure)	✅	—	—
	/system mounted rw	✅	—	—
	Root manager packages (17 packages)	✅	—	—
	su execution	✅	—	—
	Classic jailbreak paths	—	—	✅
	Modern jailbreak paths (Dopamine, palera1n)	—	—	✅
	Jailbreak URL schemes (cydia, sileo, zbra…)	—	—	✅
	fork() sandbox escape	—	—	✅
	/private/ writability	—	—	✅
	/Applications symlink	—	—	✅
Hooking	Frida port 27042	✅	—	—
	Frida process in /proc	✅	—	—
	Frida in /proc/self/maps	✅	✅	—
	Frida named anonymous mappings	—	✅	—
	XposedBridge class lookup	✅	—	—
	Xposed/LSPosed packages	✅	—	—
	XposedBridge.jar on disk	✅	—	—
	Cydia Substrate in maps	✅	✅	—
	libhooker, libsubstitute	—	—	✅
	DYLD_INSERT_LIBRARIES	—	—	✅
	Inline/PLT hook prologue scan	—	✅	—
	JNIEnv table integrity	—	✅	—
Debugger	Debug.isDebuggerConnected	✅	—	—
	FLAG_DEBUGGABLE	✅	—	—
	TracerPid ≠ 0 (Java)	✅	—	—
	TracerPid ≠ 0 (raw syscall)	—	✅	—
	Per-thread tracer check	—	✅	—
	ptrace TRACEME	—	✅	—
	Fork-based parent debugger	—	✅	—
	sysctl P_TRACED	—	—	✅
	ptrace PT_DENY_ATTACH	—	—	✅
Emulator / Simulator	Build fingerprint/model/manufacturer	✅	—	—
	QEMU system properties	✅	—	—
	Emulator phone numbers	✅	—	—
	QEMU device files	✅	—	—
	Low sensor count	✅	—	—
	TARGET_OS_SIMULATOR / env vars	—	—	✅
	Hardware model / HOME path	—	—	✅
App Integrity	APK signing cert SHA-256	✅	—	—
	Package name (repackaging)	✅	—	—
	Installer source allowlist	✅	—	—
	Bundle ID verification	—	—	✅
	Provisioning profile presence	—	—	✅
	Install source (App Store / sideload)	—	—	✅
Native Hardening	XOR-obfuscated strings	—	✅	—
	.text segment writability	—	✅	—
	Compile flags (-fstack-protector-strong, -D_FORTIFY_SOURCE=2)	—	✅	—

---

Supported Platforms

Platform	Min Version
Android	API 21 (Android 5.0)
iOS	12.0

---

Threat Model & Limitations

• Layered defense: Each threat category has multiple independent detectors. Bypassing one layer does not defeat the check.
• NDK + Kotlin: An attacker must hook both the Java and native layers simultaneously to silence all signals.
• XOR obfuscation: Sensitive path strings are encrypted in the binary; they do not appear in strings libshield.so.
• Not unbypassable: A dedicated, skilled attacker with physical device access can bypass client-side RASP. This plugin is designed to raise the cost significantly, not provide an absolute guarantee.
• Complement with server-side attestation:
  • Android: Google Play Integrity API
  • iOS: Apple App Attest / DeviceCheck

---

Legal & Ethical

• Implement a warn-and-degrade policy, not a silent brick. Show users a message; don't corrupt data.
• Respect user privacy: this plugin does not transmit any data externally.
• Implementations based on public security research: OWASP MASTG, RootBeer (Apache-2.0), published anti-tampering and Frida-detection literature.

---

Contributing

Issues and pull requests welcome. Please follow the OWASP MASTG guidelines when proposing new detection methods.

---

License

MIT