The Agent Skills Directory

[SAFE]: The skill serves as a purely instructional guide for mapping user natural language requests to specific parameters for an image transformation tool. No evidence of obfuscation, hardcoded credentials, or unauthorized network operations was found.
[COMMAND_EXECUTION]: The skill orchestrates the use of the transformation_builder tool. It provides a robust logic-based sanitization framework where the agent is explicitly instructed to rewrite vague user input into precise, predefined tool parameters rather than passing user-provided strings directly to the backend. This practice significantly reduces the risk of command or tool-specific injection attacks.

transformation-builder