The Agent Skills Directory

[SAFE]: The skill serves as a sophisticated instructional framework for AI filmmaking. It contains no executable code, obfuscated payloads, or malicious instructions.
[CREDENTIALS_UNSAFE]: The skill demonstrates safe practices for API integration by using standard placeholders (e.g., <API_KEY>) in its documentation, ensuring no real credentials are leaked or requested in an unsafe manner.
[COMMAND_EXECUTION]: The documentation provides helpful examples of how to use industry-standard CLI tools like ffmpeg and exiftool for video post-processing. These are presented as guidelines for the user's local environment and do not involve the skill executing arbitrary or unauthorized commands.
[EXTERNAL_DOWNLOADS]: The skill references official ByteDance/Volcengine API endpoints and mentions well-known creative software (e.g., Topaz Video AI, DaVinci Resolve). All external references are appropriate for the filmmaking domain and target well-known services.
[PROMPT_INJECTION]: The skill does not contain instructions that attempt to override the AI agent's core safety filters. In fact, it includes a dedicated module (seedance-copyright) and seedance-filter to help users navigate content policy restrictions and ensure their generations remain compliant with platform guidelines.

seedance-20