hunting-for-dns-based-persistence

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and ingests public, untrusted DNS data from the SecurityTrails API (see SKILL.md Step 2 and functions query_securitytrails/get_subdomains in scripts/agent.py), and that external content is parsed and used to drive detections, severities, and remediation decisions—so third-party data can materially influence the agent's workflow.