setup-cardbizpay
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from CardBizPay payment responses, which represents a potential surface for indirect prompt injection via transaction metadata.
- Ingestion points: Untrusted data enters the agent's context through the Payment Return URL and Notify URL handlers as described in
SKILL.mdand the reference files. - Boundary markers: The skill establishes a trust boundary by mandating the verification of a cryptographic
Signature(HMAC-SHA512) provided by the gateway. - Capability inventory: The skill provides instructions for performing network enquiries (
PymtEnq), transaction settlement, and refund/void operations across the Merchant and Open Payment APIs. - Sanitization: The instructions explicitly require developers to verify the signature and cross-reference critical fields such as
MerchantNo,OrderId, andTrxnAmountagainst local pending orders before updating state or fulfilling purchases.
Audit Metadata