seedance2

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly lists and requires using a web_search capability and instructs the Agent to "搜到的句式融入当前文案 / Integrate found patterns into current copy" (SKILL.md / SKILL_EN.md), and the CLI/README accept arbitrary image/video/audio URLs — meaning the Agent will fetch and interpret untrusted public web content and incorporate it into prompts and generation decisions, creating a clear indirect prompt-injection surface.