customer-billing-ops
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources such as customer emails, support complaints, and GitHub usernames, which may contain instructions intended to manipulate the agent's behavior.
- Ingestion points: External data enters the context via support complaints and customer-provided identifiers mentioned in SKILL.md.
- Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between its primary instructions and the untrusted content of customer messages.
- Capability inventory: The agent is authorized to perform billing actions such as refunds and cancellations through integrated tools like Stripe, and to generate outgoing customer communications.
- Sanitization: The instructions do not specify any validation or sanitization of the input text from support tickets before it is used for issue classification and action determination.
Audit Metadata