setup-tng-ewallet
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent on how to guide users through the processing of external payment notifications (
notifyPayment). While the skill provides strong security mitigations (signature verification), this represents an attack surface where the agent or the implemented code interacts with untrusted external data. - Ingestion points: The documentation for the
notifyPaymentcallback endpoint inSKILL.mdandreferences/tng-ewallet-api.md. - Boundary markers: The instructions explicitly warn to "verify every TNGD response and notification signature... before trusting it."
- Capability inventory: The skill enables the agent to provide logic for network API calls (Pay, Refund, Inquiry) and backend database updates for payment settlement.
- Sanitization: The skill mandates the use of RSA256 signature verification and sen-level amount/currency validation for all incoming notifications.
Audit Metadata