pilotctl

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 ingests outsider-authored free text via ~/.pilot/inbox/*.json after sending /data//summary (and even /help) to agents like list-agents and other service specialists, so a provider user can submit poison by influencing what those agents return into the inbox.