link-rot-scanner

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

• Third-party content exposure detected (high risk: 0.75). The skill fetches outsider-authored free text from the target site at runtime—specifically it retrieves {base-url}/query-index.json or {base-url}/sitemap.xml and then fetches each page’s .plain.html to extract <a href> links—so the LLM context can include untrusted page content/metadata originating from parties other than the operating user.