seo-ecommerce

Warn

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The verification logic executes a local script using the Bash tool: node ${CLAUDE_SKILL_DIR}/../../scripts/validate-jsonld.mjs --url <u>. This command employs directory traversal (../../) to execute a script located outside of the immediate skill directory structure.\n- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing untrusted external data.\n
  • Ingestion points: External HTML content is ingested via WebFetch and analyzed as rendered_dom or raw_html in SKILL.md.\n
  • Boundary markers: The instructions do not define delimiters or provide warnings to the agent to ignore instructions that might be embedded in the analyzed HTML.\n
  • Capability inventory: The skill has powerful capabilities including Bash (command execution) and WebFetch (network operations).\n
  • Sanitization: No validation or sanitization of the input HTML is described prior to its analysis.\n- [EXTERNAL_DOWNLOADS]: The skill references the Google Rich Results Test API for merchant-listing eligibility verification.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 15, 2026, 08:09 PM
Security Audit — agent-trust-hub — seo-ecommerce