The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown instructions and reference guides. It contains no executable scripts (Python, Node.js, Shell), minimizing the risk of arbitrary code execution or system compromise.
[SAFE]: Data processing is limited to reading local brand configuration files (brand-profile.md, voice.md) and processing user-provided topic descriptions. The skill does not perform network operations or access sensitive system paths (e.g., .ssh, .aws, .env).
[SAFE]: No obfuscation, prompt injection, or persistence mechanisms were detected. The skill uses standard prompt engineering techniques to ensure output quality and brand alignment without bypassing safety filters or security controls.

caption-writer