webhook-reliability

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes a workflow that reads user-pasted webhook delivery errors from the Stripe Dashboard and then performs codebase grep (e.g., searching for stripe.webhooks), which can ingest outsider-authored free text submitted by the user into the reliability/audit process.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly built to interact with Stripe (a payment gateway). It requires "Stripe MCP" via OAuth/restricted key and references specific Stripe tooling and API calls (GetMcpTools, stripe_api_search, stripe_api_details, stripe_api_read) for listing/retrieving account/payment-related data. These are specific payment gateway integration commands (not generic browser or HTTP tools) and therefore constitute direct financial execution capability.