perf-lighthouse

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md explicitly instructs running Lighthouse/LHCI against arbitrary public URLs (e.g., "lighthouse https://example.com" and the ci.collect.url entries), so the agent fetches and parses untrusted website content/JSON reports that directly drive audit results, budgets, and CI decisions.