vercel-react-best-practices
Fail
Audited by Gen Agent Trust Hub on Feb 13, 2026
Risk Level: HIGHPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [Indirect Prompt Injection] (HIGH): The skill establishes an attack surface for indirect prompt injection by instructing agents to refactor untrusted React/Next.js code. The agent's file-write capabilities combined with the absence of boundary markers create a path for malicious code comments to influence agent behavior. (Ingestion: Source files; Boundaries: None; Capabilities: File-write; Sanitization: None).\n- [Unverifiable Dependencies] (MEDIUM): Recommends the 'better-all' library from 'shuding/better-all', which is not a whitelisted trusted repository.\n- [Dynamic Execution] (LOW): Encourages the use of 'dangerouslySetInnerHTML' for inline script injection, a pattern that requires strict data control to remain safe.
Recommendations
- AI detected serious security threats
Audit Metadata