alert-triage

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs the agent to ingest “raw event(s) that triggered the alert” and other alert/telemetry fields (alert payload, SIEM/EDR/network telemetry, threat-intel lookup results) into the LLM context for triage, and those inputs are typically outsider-authored free text (e.g., SIEM/EDR-generated alert content and threat-intel reports) at runtime.