seo
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates for processing dynamic data such as blog post titles, descriptions, and slugs into page metadata and JSON-LD structured data.
- Ingestion points: Dynamic content from
post.titleandpost.descriptionis used ingenerateMetadatafunctions and JSON-LD scripts inSKILL.md. - Boundary markers: The provided code snippets do not include explicit delimiters to separate user-provided content from instructions in metadata fields.
- Capability inventory: The skill is restricted to providing JSX and TypeScript templates for metadata and SEO components; no high-privilege command execution or network exfiltration capabilities are included.
- Sanitization: The JSON-LD implementation utilizes
JSON.stringify()for serialization, providing basic protection against syntax-breaking injections, though it does not include explicit HTML sanitization for interpolated values.
Audit Metadata