The Agent Skills Directory

[SAFE]: The skill processes user-provided scene descriptions to generate structured prompts. It utilizes local safety utilities (guardRequest, safeText) to handle input and lacks any capabilities for network access, file system modifications, or arbitrary code execution, effectively mitigating prompt injection and data exposure risks.
[SAFE]: Included validation scripts (scripts/validate_output.py and scripts/validate_creative_artifacts.py) perform integrity checks on the skill's output. While these scripts utilize the subprocess module to call local helper utilities, they do not execute untrusted remote code or accept external command-line parameters that could lead to command injection.
[SAFE]: No obfuscation, hardcoded credentials, or persistence mechanisms were found in any of the skill's components. The skill's architecture relies on local relative imports and standard configuration files, adhering to safe development practices.

seedance-screenwriter