net

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly ingests untrusted remote network content — e.g., scripts/net_scan.py supports a --banner option that connects to arbitrary targets and returns service banners, and scripts/net_capture.py / scripts/net_analyze.py capture and analyze pcap traffic — so the agent will read/display third-party network payloads/banners that could contain injected instructions.