The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided text (via the --text parameter) or file content (via the --file parameter) as input for its text-to-speech engine. This content is interpolated into command-line arguments for the execution of internal scripts.
Ingestion points: User input entering the system through parameters defined in SKILL.md.
Boundary markers: Absent. The instructions do not specify any delimiters or safety warnings for the agent to apply when handling user-provided text.
Capability inventory: The skill uses subprocess calls to execute python, edge-tts, and ffmpeg via the scripts tts.py, audio_ops.py, and video_ops.py.
Sanitization: Absent. There are no explicit instructions to sanitize or validate user input before it is passed to shell-based commands.
[COMMAND_EXECUTION]: The skill operates by executing various commands and scripts through the subprocess module.
Evidence: EASEL-META.md documents the use of subprocess to call external tools such as edge-tts and ffmpeg. SKILL.md provides multiple command-line examples for synthesis, mixing, and normalization.

tts-voiceover