The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes text content from external scripts or user prompts to generate audio content, which presents a surface for instructions embedded in data.
Ingestion points: Text content in SKILL.md and scripts like VOICEOVER-SCRIPT.md processed by the /generate-voiceover workflow.
Boundary markers: No explicit delimiters or boundary instructions are provided for the text sent to the audio synthesis engine.
Capability inventory: The skill has capabilities for file system writing (save, open), network communication (ElevenLabs API), and local command execution.
Sanitization: No sanitization or input validation for the text data is described in the provided instructions.
[COMMAND_EXECUTION]: The skill requires the execution of shell commands for media processing and automation.
Evidence: Usage of ffmpeg for converting audio files and uv run tools/voiceover.py for generating scene-based audio assets.

elevenlabs