huggingface-trackio

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s autonomous agent workflow ingests outsider-authored free text at runtime by polling Trackio alerts/metrics via the trackio list alerts ... --json --since ... and related CLI “get” commands, where alert title/text are read from the alerts stored in the database/dashboards (populated by users via trackio.alert(...), potentially originating from third parties).