language-audit

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly requires a "Website URL or content set" (sitemap URL, list of page URLs, or HTML source) and the Process steps (2–5) describe parsing pages, sampling and scoring content from those public pages, so it ingests arbitrary third-party web content that directly influences audit decisions.