frontend-performance-reviewer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION] (LOW): The skill documentation includes self-authoritative safety claims (e.g., 'Verdict: Skill is bulletproof') which are flagged as an adversarial pattern intended to influence the analyzer.
- [PROMPT_INJECTION] (LOW): The skill exhibits an indirect prompt injection surface. 1. Ingestion points: External or local website content processed via Playwright browser automation (README.md). 2. Boundary markers: The skill instructions include explicit measurement blocks and user approval gates (README.md). 3. Capability inventory: Full browser automation, screenshot capture, and network analysis via Playwright. 4. Sanitization: No specific sanitization or filtering of ingested DOM content or console output is mentioned.
Audit Metadata