product-analytics

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill primarily contains documentation, platform-specific guides, and evaluation tasks for data analysis. No malicious code or hidden instructions were detected in any of the provided files.
  • [COMMAND_EXECUTION]: The evaluation tasks (e.g., in evals/merchandising-health-score-calculation/task.md) prompt an agent to create and run TypeScript scripts using npx ts-node to process sample JSON data. This is a standard part of the skill's instructional purpose for automating business reports.
  • [EXTERNAL_DOWNLOADS]: The documentation references well-known e-commerce services and apps such as Inventory Planner, Metorik, Glew.io, Hotjar, and Google Analytics. These are recognized industry-standard tools for the mentioned platforms (Shopify, WooCommerce, BigCommerce).
  • [INDIRECT_PROMPT_INJECTION]: The skill defines tasks for processing external data structures (JSON and SQL event logs). While this ingestion point is a potential surface for indirect injection from untrusted data, the skill provides specific schemas and criteria that encourage structured and validated data handling.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 12:04 PM