accounts-payable-agent
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources and other agents to execute high-impact financial transactions.
- Ingestion points: The skill accepts payment requests from other agents (Contracts Agent, Project Manager, HR) and processes external contractor invoices and recurring bills (SKILL.md).
- Boundary markers: Absent. The instructions do not define clear delimiters or provide the agent with warnings to ignore instructions embedded within the invoices or messages it processes.
- Capability inventory: The skill is capable of executing financial transfers across multiple rails, including ACH, Wire, and Crypto, using the
payments.sendtool (SKILL.md). - Sanitization: Absent. There is no evidence of content validation or sanitization for invoice metadata, memos, or reference fields which are interpolated into payment operations.
Audit Metadata