frontend-seo
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions and associated reference material consist of educational code snippets and configuration guidelines. Analysis found no evidence of prompt injection, malicious command execution, persistence mechanisms, or unauthorized data exfiltration. The code follows standard React and Next.js development patterns.
- [INDIRECT_PROMPT_INJECTION]: The skill provides a mechanism for generating page metadata and JSON-LD from application content (e.g., blog posts). This constitutes a standard data ingestion surface. 1. Ingestion points: Metadata generation logic in routes (e.g., app/blog/[slug]/page.tsx). 2. Boundary markers: Absent. 3. Capability inventory: The skill performs pure code generation and rendering without dangerous capabilities like file writing or shell execution. 4. Sanitization: Values are processed through JSON.stringify and standard React rendering, providing basic structural safety.
Audit Metadata