lighthouse-runner

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

• Third-party content exposure detected (high risk: 0.85). The required workflow ingests outsider-authored free text when the operating user supplies an arbitrary http(s)://... URL (or a local HTML file that loads external resources): Lighthouse/Chromium fetches and parses the page’s DOM/text and then the skill includes outsider-derived strings like audit.title, audit.description, and audit.displayValue in the generated text/JSON output that is fed into the agent’s LLM context.