setup-gkash
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents the handling of server-to-server callbacks (webhooks) from Gkash, which is an ingestion point for external data into the agent's context during development or debugging workflows.
- Ingestion points: Handlers for
POST <callbackurl>as described inSKILL.mdandreferences/gkash-api.md. - Boundary markers: The skill explicitly instructs the agent to treat these as trust boundaries and mandates signature verification before any state change.
- Capability inventory: The integration involves network requests (checkout, query, refund) and database state updates for fulfillment.
- Sanitization: The instructions require strict field-level matching (CID, cart ID, amount, and currency) against local records in addition to signature verification, effectively mitigating the risk of injection or spoofing.
- [EXTERNAL_DOWNLOADS]: The skill references several external resources for documentation and SDKs.
- Evidence: References to official Gkash repositories on GitHub (
github.com/gkashmy) and the official WordPress plugin registry (wordpress.org/plugins/woo-gkash/). - Assessment: These are well-known technology platforms and official vendor channels for the payment gateway. Per internal guidelines, these references are documented neutrally and do not escalate the security verdict.
Audit Metadata