content-blog
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user-provided topics and keywords for content generation.
- Ingestion points: User requests for blog posts across various platforms such as Naver, Tistory, and WordPress (SKILL.md).
- Boundary markers: Includes explicit warnings regarding the accuracy of AI-generated content and mandates a "user review" step before final publication.
- Capability inventory: Performs text generation, SEO/GEO metadata creation, and draft preparation for external publishing via the WordPress MCP connector.
- Sanitization: The workflow enforces a manual review checkpoint to prevent automated injection of malicious content into production blogs.
- [SAFE]: The skill follows professional content creation practices and references established, well-known services (WordPress, Ghost, Mailgun, Stripe) for its operational guidance. No evidence of data exfiltration, malicious command execution, or obfuscation was found.
Audit Metadata