performance
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected. The skill provides legitimate performance optimization strategies and code snippets.
- [INDIRECT_PROMPT_INJECTION]: The skill audits external URLs via performance_start_trace and lighthouse_audit, creating an inherent attack surface where untrusted content could attempt to influence agent output. This risk is neutralized by the tool-centric workflow and the skill's focus on technical Core Web Vitals metrics.
- [EXTERNAL_DOWNLOADS]: The skill references official docs and libraries from trusted vendors like Google (web-vitals, CrUX API), Mozilla, and Cloudflare for standard measurement purposes.
Audit Metadata