The Agent Skills Directory

[SAFE]: The skill is strictly designed for generating text-based JSON prompts for downstream image generation tools. It does not execute code, perform network requests, or access sensitive files.- [SAFE]: All instructions and reference materials are provided in plain text without any forms of obfuscation, such as Base64 encoding, zero-width characters, or homoglyph attacks.- [SAFE]: The skill enforces a structured JSON output format using a blueprint file, which serves as a natural constraint against common prompt injection techniques by prioritizing data structure over free-form text.- [SAFE]: No persistence mechanisms, privilege escalation attempts, or suspicious dependencies were identified in the configuration or instruction files.

prop-generation