skill-astrbot-dev

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

• Third-party content exposure detected (high risk: 0.80). The skill's docs and APIs clearly allow fetching and rendering arbitrary public web content (e.g., design_standards/visual_utils.md shows html_render(url=site_url) and html_renderer.render_t2i with network=True, and messages/components.md / visual utils also allow Image.fromURL/Record.fromURL), which means untrusted third‑party pages or user-provided URLs can be ingested and could influence agent behavior.