wcm-data-model-subscriptions-link

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documents how an agent should read and write sensitive WordPress data, creating a surface for indirect prompt injection if processing untrusted user content.
  • Ingestion points: The instructions describe accessing wc_user_membership posts, user meta keys (profile fields), and restricted REST API endpoints.
  • Boundary markers: The documentation includes clear warnings to use public PHP APIs over direct database writes to ensure proper lifecycle handling. It also highlights the security boundaries of the REST Directory endpoint.
  • Capability inventory: The skill enables the agent to interact with Custom Post Types, WordPress options (wc_memberships_rules), and user/order meta data.
  • Sanitization: The skill explicitly recommends against raw database operations, directing the agent toward object-oriented methods (e.g., $membership->update_status()) that manage side effects and data integrity.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 07:07 PM
Security Audit — agent-trust-hub — wcm-data-model-subscriptions-link