adb

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill may prompt the user for a device PIN/password and then uses commands like adb shell input text "" (or otherwise embed the PIN in adb commands) which requires the LLM to include secret values verbatim in its output.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly reads and interprets device screenshots ("Read the screenshot with the Read tool") and UI dumps (adb pull .tmp/ui.xml then "read .tmp/ui.xml") as part of its decision loop (and can open arbitrary URLs on the device), so untrusted/user-generated content shown on the device can directly influence actions and tool use.