implementing-usb-device-control-policy

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

• Third-party content exposure detected (high risk: 0.75). The required workflow includes runtime fetching of free-form JSON from a user-supplied external target URL via requests.get(...) in scripts/agent.py, which can contain outsider-authored text that is then ingested into the agent’s LLM context (directly or via the produced findings/report).

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

• Attempt to modify system services in skill instructions detected (high risk: 0.90). Flagged because the skill instructs changing system-wide configuration (GPO/MDE device-control policies and device installation restrictions) which requires administrative privileges and modifies the machine's state.