site-metadata-conventions
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest untrusted data from user requests and interpolate it into sensitive code files.
- Ingestion points: User-provided values for site title, description, and social preview images (SKILL.md).
- Boundary markers: The instructions lack explicit delimiters or warnings to ignore embedded instructions within user-provided strings.
- Capability inventory: The skill utilizes file-writing capabilities to modify app/layout.tsx, app/robots.ts, app/(replo-seo)/robots.txt/route.ts, app/sitemap.ts, and potentially public/llms.txt.
- Sanitization: No sanitization, escaping, or validation logic is specified for the user-provided metadata before it is written to the TypeScript layout file.
Audit Metadata