creator-shortlist

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

• Third-party content exposure detected (high risk: 0.85). The skill’s REQUIRED runtime workflow uses UnifAPI discovery/profile ops that fetch and ingest public web/social content authored by third parties (e.g., X community/list member profiles and their recent tweets via x/communities/{id}/members, x/lists/{id}/members, and x/users/{id}/tweets), which is outsider free text that can be included in the LLM context for scoring/evidence.