setup-omise
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references official Omise libraries and assets hosted on verified domains such as cdn.omise.co and github.com/omise. These are standard resources for payment integration.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to handle external data from webhooks and API responses (e.g., in references/omise-webhooks.md). This creates an indirect prompt injection surface where maliciously crafted data from an external source could attempt to influence the agent's logic.
- Ingestion points: External webhook POST bodies (Event objects) and API response data.
- Boundary markers: While the skill provides general integration rules, it lacks specific delimiters to separate untrusted external data within the agent's context.
- Capability inventory: The skill enables the agent to logic through charge creation, status updates, and refund operations.
- Sanitization: The skill correctly mandates HMAC-SHA256 signature verification and re-fetching data via the official Omise API to ensure authenticity and integrity.
Audit Metadata