page-seo-checker

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

• Third-party content exposure detected (high risk: 0.85). Yes: the required workflow says to use WebFetch to read the user-provided live page URL and extract/free-text elements from the returned HTML (e.g., title/meta/body/h tags/FAQ/schema), which means attacker-controlled HTML content from an outsider web page can be ingested into the agent’s LLM context during the audit.