The Agent Skills Directory

[SAFE]: No malicious patterns were identified. The skill is designed for local media processing and includes multiple safety checks and user confirmation steps for sensitive operations.
[COMMAND_EXECUTION]: The skill executes ffmpeg and ffprobe commands via the agent. It mitigates shell injection risks by explicitly instructing the agent to quote all file paths and use the -n (no-overwrite) flag to prevent data loss or accidental file replacement.
[DYNAMIC_EXECUTION]: The skill generates temporary file lists for FFmpeg's concatenation feature. It includes robust instructions for escaping single quotes within filenames and managing the -safe 0 flag, ensuring the generated lists are parsed securely by the underlying tool.
[PRIVILEGE_ESCALATION]: The included check script generates platform-specific installation commands that may include sudo. However, the skill instructions strictly prohibit the agent from executing these commands autonomously, requiring it to present the command to the user and wait for explicit approval.

modellix-ffmpeg