medusa-payments

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

• Third-party content exposure detected (high risk: 0.85). The required workflow explicitly performs runtime web-search and fetches public documentation pages (e.g., “Fetch https://docs.medusajs.com/resources/references/payment” and other site:docs.medusajs.com ... searches), and those fetched pages are outsider-authored free text that can be ingested into the agent’s LLM context.

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

• Direct money access detected (high risk: 1.00). The skill explicitly defines payment-provider interfaces and admin endpoints that perform monetary actions. It names specific gateway integrations (Stripe, PayPal) and required secrets (apiKey, webhookSecret, clientId/clientSecret). The provider API surface and service/workflow methods include initiatePayment, authorizePayment, capturePayment, refundPayment, cancelPayment, deletePayment, getPaymentStatus, updatePayment, retrievePayment and admin routes like POST /admin/payments/:id/capture and POST /admin/payments/:id/refund. These are specific payment gateway operations intended to move funds or reverse transactions, not generic tooling. Therefore this skill grants direct financial execution capability.