protocol-reverse-engineering

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

• Third-party content exposure detected (high risk: 0.80). The skill ingests and analyzes captured network traffic and intercepted HTTP/HTTPS responses from arbitrary hosts (e.g., mitmproxy interception, Wireshark "Export Objects"/tshark -Y "http" -> http_traffic.json, and reading capture.pcap with scapy), which can contain untrusted third‑party content that might carry indirect prompt injections.