The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes local project files to determine video edit ranges, creating an attack surface where untrusted data could influence command execution.
Ingestion points: The skill reads edit metadata from ./public/<slug>.edit.json.
Boundary markers: None are used to isolate the data.
Capability inventory: The skill utilizes node, ffmpeg, and curl for file processing and network communication.
Sanitization: Data from the JSON file is interpolated directly into a node -e shell command without validation.
[DYNAMIC_EXECUTION]: The skill uses node -e to execute an inline JavaScript snippet that parses edit ranges and generates filter strings for ffmpeg.
[EXTERNAL_DOWNLOADS]: The skill interacts with ElevenLabs' official API for video dubbing and translation services.

video-translate