The Agent Skills Directory

[SAFE]: The skill consists of natural language instructions guiding an AI to generate image prompts following specific artistic aesthetics (Hasselblad, Leica, Ricoh GR). It does not contain executable code, shell scripts, or malicious logic.
[DATA_EXFILTRATION]: No sensitive file access or network communication was detected. The skill instructs the agent to save files using a configurable <OUTPUT_ROOT> path, which is a standard safety measure for managing generated assets without interfering with system directories.
[PROMPT_INJECTION]: The instructions include directives to hide internal prompt details and鏡头计划 (lens plans) from the user. These are operational formatting constraints designed to improve user experience rather than attempts to bypass platform safety filters or extract system prompts.
[SAFE]: The skill processes user-provided locations and topics for image generation. While these are external ingestion points, the capability is limited to image model interpolation with no evidence of dangerous operations (like file writing or command execution) being triggered by this data.

fantasy-photography-simulation