shopify-expert
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external sources.\n
- Ingestion points: Processes product metadata, customer inputs, and storefront data via GraphQL queries and Liquid templates (documented in
SKILL.mdandreferences/storefront-api.md).\n - Boundary markers: Lacks explicit instructions to the agent to distinguish between data and instructions when processing external store content.\n
- Capability inventory: The skill utilizes the Shopify CLI to perform administrative actions like
shopify theme pushandshopify app deploy(documented inSKILL.md).\n - Sanitization: Relies on default output encoding in Liquid and modern web frameworks, but does not implement specific sanitization for agent-level instruction processing.\n- [COMMAND_EXECUTION]: The skill extensively uses the Shopify CLI for theme management, app deployment, and development workflows, including
shopify theme devandshopify app dev(found inSKILL.md).\n- [EXTERNAL_DOWNLOADS]: The skill references and downloads official tools and configurations from well-known technology services.\n - Fetches official Shopify app and Hydrogen scaffolds from the public npm registry using
npm create @shopify/app@latest.\n - Recommends the installation of Lighthouse CI for performance auditing via
npm install -g @lhci/cli.\n - Provides examples for loading third-party scripts, which is standard for e-commerce integrations.
Audit Metadata