The Agent Skills Directory

[SAFE]: The skill operates entirely offline and implements legitimate design token definitions and mathematical calculations aligned with the author context.
[PROMPT_INJECTION]: No behavioral override patterns, ignore instructions, or jailbreak techniques were found in the text or metadata.
[DATA_EXFILTRATION]: There are no indicators of data exposure, credential harvesting, or network calls to external sites. The URL referenced is for standard web feature verification.
[DYNAMIC_EXECUTION]: Code components use strict type definitions and explicit mathematical logic. No unsafe deserialization or runtime evaluation of arbitrary strings is present.
[COMMAND_EXECUTION]: Shell commands and script targets are confined to validating local files and running the static token generator tool.

spacing-ladder