frontend-react-performance
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied React and Next.js code, which introduces a potential surface for indirect prompt injection attacks.
- Ingestion points: Untrusted source code enters the agent context via the
$ARGUMENTSplaceholder at the end ofSKILL.md. - Boundary markers: The instructions lack explicit delimiters (e.g., XML tags or triple backticks with 'ignore' warnings) to prevent the agent from following instructions that might be embedded in the analyzed code.
- Capability inventory: The skill allows for code review, debugging, and refactoring, which involves reading files and generating code changes to the project.
- Sanitization: No sanitization, validation, or filtering of the external code content is defined in the instructions.
- [SAFE]: The skill metadata references Vercel's React Best Practices, which is an official resource from a well-known service provider.
Audit Metadata