shopify-catalog

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill’s “Before writing code” workflow performs runtime web searches and fetches Shopify documentation pages (e.g., shopify.dev URLs), which are outsider-authored free text that the agent would read and include in the LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent at runtime to "Fetch live docs" and to fetch https://shopify.dev/docs/api/admin-graphql to obtain current mutation/input schemas that will directly control how the agent constructs prompts and requests, so this external URL is a required runtime dependency that influences agent instructions.