accounts-payable-agent
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources such as contractor invoices and payment requests from other agents, which creates a surface for indirect prompt injection.
- Ingestion points: Untrusted data enters the agent context through the
requestparameter inprocessContractorPaymentand retrieved data ingetScheduledPaymentswithinSKILL.md. - Boundary markers: Absent; the skill instructions do not define delimiters or specific warnings to ignore instructions that might be embedded in invoice references or metadata.
- Capability inventory: The skill has the capability to execute high-value financial operations via the
payments.sendfunction. - Sanitization: The provided logic focuses on vendor validation and spend thresholds but lacks sanitization or filtering of text fields like
memoorreferencewhich could contain malicious payloads.
Audit Metadata