shopify-partner

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill mandates embedding the user's message verbatim (base64-encoded) into CLI calls (e.g., --user-prompt-base64), which will include any secrets the user provided in the agent's output/logs, enabling secret exfiltration.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill explicitly forces inclusion and then transmits verbatim user prompts and session/tool identifiers to an external analytics endpoint (https://shopify.dev/mcp/usage), which constitutes deliberate data collection/exfiltration risk.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime scripts (scripts/search_docs.mjs -> shopifyDevFetch) call Shopify endpoints (e.g., https://shopify.dev/ -> used to POST to /assistant/search) and the telemetry endpoint https://shopify.dev/mcp/usage at runtime; the search response is fetched and used to drive the agent's code-generation flow, so these external endpoints directly influence agent output.