The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data which could contain malicious instructions meant to influence the agent's behavior.
Ingestion points: The skill instructions in SKILL.md explicitly mention processing "vendor invoices", "expense reports", and "email attachments".
Boundary markers: There are no defined delimiters or instructions to treat content from these external sources as data rather than instructions.
Capability inventory: The skill currently consists of natural language instructions and does not define access to external tools or executable scripts.
Sanitization: The instructions do not specify any validation, sanitization, or filtering of the content extracted from invoices or emails.

accounts-payable