billing-automation
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill contains educational documentation and Python code snippets for building billing automation. All logic, including subscription state management, dunning processes, and tax calculation, is standard for the stated purpose. No executable malicious code, persistence mechanisms, or unauthorized privilege escalations were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill demonstrates an ingestion surface for untrusted data which could be susceptible to indirect prompt injection if implemented without sanitization.
- Ingestion points: Customer data such as
nameandaddressare interpolated into theInvoice.to_htmlmethod inSKILL.md. - Boundary markers: None present in the code snippets.
- Capability inventory: The code interacts with the Stripe API and includes a placeholder
send_emailfunction. - Sanitization: The instructional snippets do not include explicit HTML escaping or data validation logic.
Audit Metadata