hunt-metrics-exposure

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill contains explicit reconnaissance and data-exfiltration guidance (probing /metrics and /actuator endpoints to harvest env/config and AI/model usage) and directs operators to pivot to exploitation (prompt injection, DoS), indicating intentional malicious intent.

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow fetches free-text from attacker-controlled target endpoints (e.g., via curl of /metrics, /actuator/health, /actuator/metrics, and /telescope/requests into files that are then grepped/parsed), so outsider-authored content can be ingested if an outsider can provide/post the HTTP responses or influence the target.