Skip to content

Security: dongnh311/DeviceGuardSDK

Security

SECURITY.md

Security Policy

DeviceGuard SDK is a client-side device-security library. We take vulnerabilities in the detection logic, false-negative gaps, and any issue that could compromise consumer applications seriously — please follow the process below to report them privately.

Reporting a vulnerability

Do not file a public GitHub issue for security problems.

Use one of the private channels instead:

  • GitHub Security Advisories (preferred) — on the repo, Security tab → Report a vulnerability. This opens a private advisory visible only to the reporter and the maintainers, with a dedicated thread for coordinated disclosure.
  • Emailhoaidongit5@gmail.com. Include [DeviceGuard SDK Security] in the subject. If you need to send an encrypted payload, ask for a PGP key in your first message.

When reporting, please include:

  1. The affected module (deviceguard-core, deviceguard-rootcheck, …) and version / git SHA you tested against.
  2. The platform (Android + API level, iOS version, JVM version, or browser).
  3. A concrete reproduction — either a minimal code snippet, or a description of the device state that triggers the issue.
  4. The impact you believe the issue has: bypass of a detection, information leak, crash / DoS, signature-verification weakness, etc.

What we commit to

  • Acknowledgement within 5 business days of receipt.
  • Triage and severity classification within 10 business days, shared back with the reporter.
  • Fix timeline discussed openly in the advisory thread. For high-severity issues we aim for a coordinated release within 30 days.
  • Credit to the reporter in the release notes and CHANGELOG.md, unless the reporter prefers to remain anonymous.

What's in scope

  • Bypasses of any DeviceGuard detection (root / jailbreak, emulator, app integrity, VPN / proxy) that defeat the SDK without reasonable effort.
  • Information disclosure — any fingerprint signal that leaks PII we document as non-collected.
  • Signature / certificate comparison flaws that allow a tampered app to pass integrity checks.
  • Logic issues in the risk-scoring or report-assembly code that could mis-classify a genuinely compromised device as safe.

What's out of scope

  • Theoretical bypasses that require root + Frida + hook framework + deep reverse engineering. A determined attacker with kernel-level control (Magisk-hide, Frida-gadget injection, custom LSM patches) can defeat any client-side detection. This is a known limitation of every client-side security SDK and not in itself a vulnerability — we treat it as a floor to be raised by layering server-side attestation on top of the client signals DeviceGuard surfaces.
  • Issues in server-side attestation (Play Integrity API, SafetyNet, Apple DeviceCheck) — DeviceGuard does not wrap these; it is your responsibility to correlate server-side.
  • False positives on niche device configurations. Open a regular GitHub issue with the device / OS details and we'll tune weights.
  • Issues in dependencies (kotlinx-coroutines, serialization, kotlincrypto) — report upstream, optionally file a tracking issue here once the upstream has a fix.

After a fix ships

A public advisory is published via the GitHub Security Advisories feature with a CVE if warranted, and the CHANGELOG.md entry for the release includes a Security subsection linking to the advisory.

There aren’t any published security advisories