networking

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

• Third-party content exposure detected (high risk: 0.80). The skill's troubleshooting workflow and example commands (e.g., dig @8.8.8.8, curl -v http://remote_host:8080/health, traceroute/mtr, and docker run --rm ... curlimages/curl ...) explicitly instruct fetching and interpreting responses from arbitrary external hosts (example.com, remote_host, public DNS servers), which are untrusted third‑party sources.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs running sudo commands that modify system state (firewall rules via iptables/nft/ufw, /etc/hosts, routes, flushing resolvectl caches, etc.), which can change or compromise the machine and potentially lock out access, so it should be flagged.