shopify-mcp-server

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's docs and examples explicitly show placing API secrets and tokens into command-line arguments, JSON config, environment variables, and inline-start commands, which requires an LLM to output or embed secret values verbatim (high exfiltration risk).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill documentation includes commands that fetch and run remote code at install/run time — notably "git clone https://github.com/Cesarjoquin/shopify-mcp.git" and the "npx shopify-mcp" invocation which will download and execute the package — so external code is fetched and executed during setup/runtime.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill is a Shopify Admin MCP server exposing specific order-management tools and scopes. The docs explicitly list order actions including "refund" and provide examples and scopes (e.g., write_orders) plus access tokens required to authenticate. Those items indicate the ability to perform payment-related operations (issue refunds / modify orders) via the Shopify API—i.e., direct financial execution capability.