billing-automation
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides code patterns in
references/details.mdthat interpolate untrusted data into HTML templates without sanitization, creating a vulnerability surface. - Ingestion points: Customer-supplied data such as
customer_nameandcustomer_addressare used in theInvoiceclass. - Boundary markers: None present to distinguish between instructions and data within the template.
- Capability inventory: The
to_htmlmethod generates raw HTML strings that could be rendered by downstream systems or the agent. - Sanitization: The implementation uses Python's
.format()method for string interpolation without escaping or validating the HTML content, which is a common vector for injection attacks.
Audit Metadata