android-mobile-security-sandbox-jamboree

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill includes Frida/Objection hooks and example scripts that explicitly capture, log, and write secrets (tokens, keys, plaintext) verbatim (e.g., console.log of tokens, writing extracted-token.txt), which requires the agent to handle and output secret values directly.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). Mixed risk: while ara.so, example.com and the local 192.168.1.100:8080 address are not inherently suspicious, the GitHub repo (hero-mike) is an unknown personal repository that distributes binaries/APKs and scripts that require elevated privileges, and the frida releases URL (a direct release/download path) can be used to host executables — together these are potentially dangerous and warrant caution.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill contains multiple deliberate high-risk, dual-use capabilities: explicit credential/data exfiltration (Frida hooks logging plaintext/keys and dumping tokens to /sdcard), persistent system compromise and hiding (Magisk modules, init scripts, remounting/replacing system libraries, modifying build.prop, installing system CA certs), and remote code injection/runtime manipulation (Frida/Objection spawn/attach and dynamic hook deployment) — all of which can be readily abused as backdoors or for theft.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly clones and then executes the repository (git clone https://github.com/hero-mike/Android-Mobile-Security-Sandbox-Testing.git followed by ./orchestration/deploy.sh) and also fetches/installs Frida server via wget from https://github.com/frida/frida/releases/... which downloads and runs remote binaries — both actions fetch remote code that is executed at runtime.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). This skill explicitly instructs obtaining and using root/su (adb root, adb shell su -c), remounting and modifying /system files (installing CA certs, editing build.prop), installing Magisk modules, changing iptables, and other privileged actions that modify the device/emulator system state and bypass security mechanisms.