agent-ux-optimizer

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

• Third-party content exposure detected (high risk: 1.00). The skill's required Testing Workflow Phase 2 explicitly instructs running "npx lighthouse " (and the imported tool list includes WebFetch/WebSearch), which fetches and analyzes arbitrary public URLs/pages—untrusted third-party content that the agent must read and that can materially influence its recommendations.