The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines patterns for ingesting and processing untrusted data from web requests.
Ingestion points: The save_product_meta method in SKILL.md processes $_POST data, and the get_analytics_summary method processes REST API parameters.
Boundary markers: Nonce verification (wp_verify_nonce) is used to validate the authenticity of the request.
Capability inventory: The skill demonstrates database write capabilities via update_post_meta and update_meta_data, data retrieval via wc_get_orders, and potential network operations via placeholder fulfillment logic.
Sanitization: The code correctly uses sanitize_text_field() and absint() to sanitize external inputs before they are used in business logic or stored in the database.

woocommerce-plugin-development