frontend-performance

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to run audits on target URLs and analyze network requests/third-party scripts (see Steps 3, 10, and 13 in SKILL.md), which requires fetching and interpreting content from public/untrusted third-party origins (analytics, ads, embeds) and using those findings to drive tool use and remediation decisions.