signals-scout-logs

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

• Third-party content exposure detected (medium risk: 0.65). This skill’s required runtime workflow reads log/message content via query-logs (and uses logs-services-create service/error-rate lists), which are generated by other parties/services and may include outsider-authored free text (e.g., log message bodies containing user input) fed into the agent’s LLM context for analysis.