shopify-onboarding-merchant

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • Indirect Prompt Injection Surface: The skill processes user-supplied CSV files to facilitate product migration from various commerce platforms.
  • Ingestion points: Merchant-provided CSV files are read in the "Import products from another platform" section (SKILL.md).
  • Boundary markers: The skill uses structured JSON construction for GraphQL mutations, though it does not explicitly define delimiters to separate instructions from data within CSV cells.
  • Capability inventory: The skill performs command execution via shopify store execute (including mutations), shopify store auth, and package installation via npm and brew.
  • Sanitization: The skill includes validation steps (checking variant and option limits) and mandates writing variables to a temporary JSON file to prevent shell injection, which serves as a significant security mitigation.
  • External Package Installation: The skill automates the installation of the Shopify CLI (@shopify/cli) using standard package managers. These downloads target official, well-known service repositories and are necessary for the skill's core functionality.
  • Command Execution for Store Management: The skill executes several shell commands to authenticate with Shopify and manage store data. It follows security best practices by avoiding the direct inlining of merchant-supplied strings into shell commands, opting instead to pass data through temporary JSON files to mitigate command injection risks.
  • Authentication Flow: The skill triggers an OAuth flow using the Shopify CLI, which involves starting a local callback server. This is a standard procedure for command-line authentication with the platform.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 05:50 PM
Security Audit — agent-trust-hub — shopify-onboarding-merchant