seo-page

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md workflow explicitly instructs the agent to "Fetch page (URL via WebFetch, or read local file)" and then parse and analyze the fetched page (meta, schema, headings, and an "LLM-readiness" score), so arbitrary public web pages could supply untrusted, user-generated content that the agent reads and uses to drive its analysis and decisions.