schema
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user prompts and local context files to generate JSON-LD markup, which represents an indirect prompt injection surface.
- Ingestion points: The skill reads website content, CMS data, and product marketing context files (such as .agents/product-marketing.md) provided by the user to populate schema fields.
- Boundary markers: The skill uses markdown code blocks to delimit the generated JSON-LD output in its responses.
- Capability inventory: The skill generates JSON-LD code snippets for the user to implement but does not perform automated file-writes, execute shell commands, or make network requests itself.
- Sanitization: The skill provides proactive security guidance in its reference materials and evaluation criteria, instructing users to escape the less-than character as \u003c when embedding generated JSON-LD in HTML script tags to prevent cross-site scripting (XSS) and script injection.
Audit Metadata