video-extract
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill utilizes ffmpeg and ffprobe, which are standard and trusted open-source utilities for media processing. The instructions recommend installation via Homebrew, a well-known package manager.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied video files which represents an indirect prompt injection surface if the agent processes the resulting metadata (e.g., via ffprobe).
- Ingestion points: External media files provided by the user (e.g.,
input.mp4). - Boundary markers: None explicitly defined for command output processing.
- Capability inventory: Bash execution for
ffmpegandffprobeoperations, including file read and write capabilities. - Sanitization: There is no explicit sanitization of media metadata or filenames mentioned in the instructions.
Audit Metadata