wds-5-agentic-development

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly instructs agents to use Puppeteer to navigate to and read live/public page URLs (see INLINE-TESTING-GUIDE.md and SEO-VALIDATION-GUIDE.md with page.goto and checks of meta/heading/structured-data), meaning it ingests untrusted third‑party web content and uses those readings to decide fixes or next actions.