setup-shopeepay-gateway
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides detailed instructions for processing external webhook data (Notify Transaction Status callbacks), creating an attack surface for indirect prompt injection.
- Ingestion points: The callback handler is designed to ingest raw bytes from an external POST request to a merchant-supplied URL.
- Boundary markers: The skill instructions mandate the use of cryptographic boundary markers by verifying the
X-Airpay-Req-HHMAC-SHA256 signature before parsing the payload. - Capability inventory: The suggested implementation involves querying a database to match orders and potentially initiating network requests for transaction status re-verification.
- Sanitization: The skill emphasizes strict sanitization via constant-time signature comparison and cross-referencing the
amountandpayment_reference_idwith internal state, which significantly mitigates injection risks. - [DATA_EXPOSURE]: The documentation includes a sample secret key and a cryptographic test vector.
- Evidence: In
references/shopeepay-gateway-api.md, a sample secret keypz148x0gXyPCLHxnlhEydNLg55jni91iand a corresponding Base64 HMAC digest are provided for unit testing purposes. These are explicitly identified as non-production samples and follow standard industry documentation practices.
Audit Metadata