intercepting-mobile-traffic-with-burpsuite

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

• Third-party content exposure detected (high risk: 0.75). Yes—runtime-captured traffic from the target app (outsider-authored content in the sense of non-user-provided request/response bodies) is ingested by the agent via parsing user-supplied HAR/XML files and then placed into the agent’s LLM context through generated report text (e.g., request/response bodies and extracted findings).

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

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill explicitly instructs modifying device system files (adb push/cp to /system/etc/security/cacerts/, chmod), using Magisk/root and iptables, and running Frida/Objection to bypass certificate pinning — all actions that alter system state or bypass security controls on a target device.