canvas-design

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly accepts WeChat public-account (微信公众号) article content from users and directly passes that user-provided, public/user-generated text into scripts/generate_cover.py as the image-generation prompt, so it ingests arbitrary third-party content that could carry indirect prompt injection.