moonpay-commerce
Pass
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill leverages the
mpCLI tool to execute store listing, product search, and cart management functions. - [DATA_EXFILTRATION]: The
checkoutcommand processes sensitive personally identifiable information (PII), including name, email, and physical address, which is necessary for order fulfillment by the vendor. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and displays product descriptions from external Shopify stores. Evidence Chain: 1. Ingestion points: Product titles and descriptions fetched during
product searchandretrievecommands (SKILL.md). 2. Boundary markers: No specific delimiters or safety instructions are defined for the fetched content. 3. Capability inventory: Use of thempCLI tool for transactions and data display. 4. Sanitization: No sanitization or filtering of external store content is mentioned in the skill definition.
Audit Metadata