wordpress-woocommerce-development
Pass
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains multiple PHP code snippets that facilitate indirect prompt injection by interpolating untrusted data into prompts sent to the WordPress 7.0 Abilities/AI API.
- Ingestion points: Untrusted data enters the agent context through several functions in
SKILL.md:generate_ai_product_description(product title and short description),ai_check_order_fraud(shipping address, billing email, and order total),handle_ai_product_question(user-submitted questions via AJAX), andai_validate_order(checkout fields like email, phone, and address). - Boundary markers: The code snippets do not implement any boundary markers or delimiters (like XML tags or clear separators) to prevent the AI from interpreting the data as instructions.
- Capability inventory: The skill includes capabilities to automate product description generation, process order fraud analysis, and provide automated customer service responses based on the results of these AI prompts.
- Sanitization: While standard WordPress sanitization functions like
sanitize_text_fieldandsanitize_emailare used, these are designed to prevent XSS and database injection, not prompt injection. There is no evidence of filtering or escaping designed to neutralize adversarial prompt content within the data strings.
Audit Metadata