plan-launch

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

• Third-party content exposure detected (medium risk: 0.30). The skill “Plan Launch” includes runtime activities that would monitor/ingest outsider-authored free text such as replying to every App Store review (and also reading public comments on Reddit/Twitter as part of engagement), which requires LLM consumption of external user text without necessarily selecting a specific beforehand-approved item.