social-content

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

• Third-party content exposure detected (high risk: 0.80). SKILL.md's "Reverse Engineering Viral Content" workflow explicitly instructs the agent to "Find creators" and "Collect data — 500+ posts" from public social platforms (LinkedIn, Twitter/X, Instagram, TikTok, Facebook), so the skill requires ingesting and analyzing untrusted, user-generated third‑party content that can influence decisions and actions.