product-information-management
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a data ingestion pipeline from external PIM APIs (Akeneo and Salsify), which creates an attack surface for indirect prompt injection if the external data source contains malicious instructions.
- Ingestion points: Akeneo REST API (SKILL.md) and Salsify API (evals/salsify-integration-and-data-quality-rep/task.md).
- Boundary markers: Absent. The instructions do not recommend using delimiters or specific "ignore embedded instructions" warnings when interpolating external product data into storefront objects.
- Capability inventory: The skill code performs network requests (fetch), database writes (upserts), and triggers cache revalidation or shell commands based on the ingested data.
- Sanitization: While the logic maps and transforms data into a flat structure, it lacks explicit sanitization or validation to prevent the execution of malicious payloads (e.g., HTML or Markdown injection) by downstream storefront components.
- [COMMAND_EXECUTION]: The WooCommerce integration instructions describe using the WP-CLI command
wp transient delete-allto clear the WordPress cache after a successful product sync. - Execution method: Instructs the use of shell commands to manage system or application state.
- Context: While this is a standard administrative task for WordPress developers, it involves the execution of external binaries based on the skill's operational flow.
Audit Metadata