shopify-ai-toolkit-wrapper

Warn

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: MEDIUMPERSISTENCEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PERSISTENCE]: The skill provides instructions in references/mcp-config.md to append environment variable exports to the user's ~/.bashrc file. This modification of a shell initialization script is a persistence mechanism used to maintain configuration across sessions.
  • [EXTERNAL_DOWNLOADS]: The skill configuration suggests using npx to fetch and execute the @shopify/ai-toolkit-mcp package from the official Shopify organization scope on the NPM registry. This is a reference to software from a well-known service.
  • [INDIRECT_PROMPT_INJECTION]: The skill enables the agent to search external Shopify documentation, creating a vulnerability surface for indirect prompt injection.
  • Ingestion points: External Shopify documentation content accessed via the MCP search tool (mentioned in SKILL.md).
  • Boundary markers: The skill does not define specific delimiters or instructions to isolate the retrieved documentation content from the agent's internal instructions.
  • Capability inventory: The skill possesses tools for file manipulation (Read, Write, Edit), shell command execution (Bash restricted to npm), and metadata grep.
  • Sanitization: No data validation or content sanitization logic is described in the provided workflows.
  • [COMMAND_EXECUTION]: The skill instructions involve executing shell commands via the Bash tool, specifically npx and npm, to configure the MCP server and troubleshoot connectivity.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 9, 2026, 03:47 AM
Security Audit — agent-trust-hub — shopify-ai-toolkit-wrapper