The Agent Skills Directory

[SAFE]: No prompt injection or jailbreak attempts detected. The instructions follow standard agent guidance for the task of document generation.
[SAFE]: No data exfiltration or credential harvesting patterns identified. The collection of business and payment details is restricted to the user's manual input for the purpose of populating the generated invoice.
[SAFE]: No external downloads or remote code execution risks found. The skill leverages Model Context Protocol (MCP) tools for document processing which are consistent with its stated purpose.
[SAFE]: No obfuscation, zero-width characters, or hidden content detected in the skill metadata or body text.
[SAFE]: The skill presents a standard indirect prompt injection surface as it ingests untrusted client data, but this is inherent to its primary purpose and includes no exploitable code paths in the provided files.

invoice-generator