The Agent Skills Directory

[SAFE]: The skill is designed for the legitimate purpose of generating cinematic prompts and production documents. It follows a structured workflow to transform user input into a markdown file using internal templates and reference materials.
[DATA_EXFILTRATION]: No network-enabled tools or commands (such as curl, wget, or fetch) are utilized. The skill does not attempt to access sensitive system paths or credentials.
[COMMAND_EXECUTION]: The skill uses a 'Write' tool to save its generated output to a file, which is consistent with its stated purpose. There are no patterns of arbitrary shell command execution, privilege escalation, or persistence mechanisms.
[PROMPT_INJECTION]: The instructions provide clear boundaries for creative generation and include self-checks for quality and consistency. No malicious bypass patterns or 'DAN' style injections were detected.
[OBFUSCATION]: The skill's content and its extensive reference libraries (movie styles, camera language, etc.) are written in clear, plain text. No hidden URLs, encoded payloads, or steganographic techniques were found.

cinematic-prompt