The Agent Skills Directory

[SAFE]: The security audit of the skill reveals no malicious patterns, obfuscation, or unauthorized data access. The skill operates within its intended scope of document generation.
[EXTERNAL_DOWNLOADS]: The skill communicates with the DocGen Engine API located at https://docgen-production-503d.up.railway.app. This network activity is a core functional requirement for generating PDF documents from user-provided data.
[COMMAND_EXECUTION]: The skill is designed to save binary PDF data received from the API to the local file system. This is the primary method of delivery for the requested document.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it processes untrusted user data and scans external PDF files to extract document fields.
Ingestion points: User-provided invoice details and data extracted from uploaded PDFs via the /api/v1/scan endpoint (documented in references/api.md).
Boundary markers: None identified; user data is interpolated into API requests without explicit delimiters.
Capability inventory: The skill can perform network POST requests and write binary files to the local disk.
Sanitization: There is no evidence of input sanitization or validation before data is processed by the API.

docgen-generate