lov-image-creator

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s mechanism selection and inputs ultimately pass user-supplied free text as the prompt argument to gen_image.py (Mechanism 1) which then calls ZenMux/Gemini with contents=[prompt], so outsider-authored text is ingested directly at runtime.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill's render_to_png runtime (scripts/render_to_png.py loads the generated HTML with Playwright) will cause the page to fetch and execute third-party scripts from CDNs (e.g., https://cdn.jsdelivr.net/npm/react@19/umd/react.production.min.js, https://cdn.jsdelivr.net/npm/react-dom@19/umd/react-dom.production.min.js, https://cdn.tailwindcss.com, https://fonts.googleapis.com/css2?family=Noto+Sans+SC:wght@300;400;500;700;900&family=Noto+Serif+SC:wght@400;700&display=swap, https://cdn.jsdelivr.net/npm/@babel/standalone/babel.min.js) at runtime, so these external URLs are required dependencies that execute remote code in the rendering context.