stream-aggregation-helper

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

• Third-party content exposure detected (low risk: 0.10). The skill only reads trusted, first-party text from its own prompt/docs (this SKILL.md) and at runtime it performs user-driven explicit metric queries against the user’s VictoriaMetrics URL (e.g., /api/v1/status/tsdb and /api/v1/query) rather than ingesting arbitrary outsider-authored free text from an always-on feed/queue without selecting a specific metric.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The skill instructs editing and placing vmagent config files (e.g., /etc/vmagent/...), updating VMAgent CRDs, and sending signals/reloads to the vmagent process — all actions that change system or cluster state and typically require elevated privileges.