blog-taxonomy
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes local files to suggest tags, which serves as an ingestion point for untrusted content.
- Ingestion points: Content is read from local files during the suggestion and audit workflows in
SKILL.md. - Boundary markers: Not explicitly defined for the file content being parsed.
- Capability inventory: Authenticated API interactions (GET, POST, GraphQL) with WordPress, Shopify, Ghost, Strapi, and Sanity.
- Sanitization: The skill extracts headings and analyzes word frequency but does not explicitly describe sanitization of input text for malicious instructions.
- [COMMAND_EXECUTION]: The skill performs network operations to interact with external CMS APIs.
- Evidence: Implements adapters for multiple CMS platforms requiring network access to
CMS_URL. - Safety Controls: The skill documents significant security mitigations, including mandatory HTTPS, blocking of private and reserved IP addresses (SSRF protection), and an optional host allowlist (
CMS_ALLOWED_HOSTS).
Audit Metadata