qa-detecting-visual-regressions

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

• Third-party content exposure detected (high risk: 0.90). The skill's capture.py explicitly navigates to arbitrary base_url + page paths (e.g., --url and --pages in SKILL.md and scripts/capture.py) and ingests live web pages/screenshots which are then analyzed by scripts/compare.py to drive regressions/classification, so untrusted third-party web content can directly influence the tool's decisions and outputs.