design-qa-review-skill

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s REQUIRED runtime workflow can ingest outsider-authored free text when it performs URL-based page analysis (e.g., --page-url / pageUrl), since it may fetch arbitrary public/staging pages and then use their readable content (DOM/text) as LLM context for analysis/reporting, enabling indirect prompt injection.