The Agent Skills Directory

[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes external data from invoices and vendor records. While this is an inherent risk for its function, the absence of boundary markers or sanitization logic in the workflow examples is noted.
Ingestion points: getScheduledPayments and processContractorPayment accept external data (SKILL.md).
Boundary markers: None present to delimit external data from instructions.
Capability inventory: Access to financial tools like payments.send and payments.getHistory (SKILL.md).
Sanitization: No explicit input validation or escaping for invoice metadata is defined in the examples.
[NO_CODE]: No executable code or scripts are included in this skill; it consists of high-level instructions and configuration metadata.

agency-accounts-payable-agent