meta-ads-creative

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

• Third-party content exposure detected (high risk: 0.90). The skill's required Phase 1 Research workflow (e.g., "Step 2: Reddit/Community Research" in SKILL.md and references/creative-research-methods.md) explicitly instructs scraping and ingesting user-generated public sources such as Reddit, Facebook groups, parenting forums, and competitor ad libraries, which the agent is expected to read and use to drive ad decisions and creative actions, exposing it to untrusted third‑party content that could enable indirect prompt injection.