demoz-skill

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

• Third-party content exposure detected (low risk: 0.10). At runtime, the workflow reads outsider-authored free text only as user-provided strings passed into video/audio generation scripts (e.g., --prompt/--text), but this is not shown to be ingested by the in-session brand-extraction agent/LLM (which instead reads brand media files locally and writes a JSON candidate that is then deterministically WCAG-gated).