mpp-stripe-method

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

• Third-party content exposure detected (high risk: 0.85). The REQUIRED workflow includes runtime web-fetching of public Stripe documentation pages (e.g., https://docs.stripe.com/... and web-search results), and those pages’ readable text would be ingested into the agent’s LLM context, which is outsider-authored free text from the public web.

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

• Direct money access detected (high risk: 1.00). The skill explicitly defines Stripe payment operations: agent-side creation of Shared Payment Tokens via a Stripe API endpoint (POST /v1/test_helpers/shared_payment/granted_tokens) and server-side consumption of those tokens by creating/confirming PaymentIntents (POST /v1/payment_intents with shared_payment_granted_token). It includes concrete API parameters, configuration (mppx stripe.charge with secret keys), and a full payment flow. These are specific payment-gateway actions that enable an agent to initiate and complete fiat/card charges, i.e., direct financial execution.