seo-meta-tags
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection as it instructions the agent to ingest untrusted user data (e.g., blog content, CMS data, API responses) to populate metadata templates.
- Ingestion points: The skill utilizes
{{PLACEHOLDER}}tokens across its templates (HTML, Next.js, React, Vue) and handles dynamic data in its reference components. - Boundary markers: The skill provides instructional boundaries by telling the agent to "always escape HTML entities" and "never leave placeholders in production output."
- Capability inventory: The skill empowers the agent to create and modify critical application files (index.html, metadata.ts, SEO.tsx) which can influence how a site is indexed and rendered.
- Sanitization: The skill contains a dedicated 'Security' section requiring the use of
JSON.stringify()for structured data and recommending the use of sanitization libraries like DOMPurify or sanitize-html for any dynamic HTML rendering.
Audit Metadata