audio-edit
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines workflows for processing external media files (video and audio) using
ffmpeg. This represents a data ingestion surface where the agent acts on potentially untrusted user-supplied content. - Ingestion points: Input files (
in.mp4,bg.mp3) specified in theffmpegcommand templates withinSKILL.md. - Boundary markers: None; the instructions assume direct interpolation of file paths into shell commands.
- Capability inventory: Shell command execution using the
ffmpegutility for media encoding and filtering. - Sanitization: No explicit validation or sanitization of the input media files is included in the instructions.
Audit Metadata