video-editing
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: Includes standard FFmpeg and shell commands for video manipulation. While it uses the
-safe 0flag for concatenation, this is a common practice for handling varied file paths in production environments. - [EXTERNAL_DOWNLOADS]: References well-known and trusted platforms such as ElevenLabs, fal.ai, and Remotion for media processing and asset generation.
- [PROMPT_INJECTION]: Evaluated for indirect prompt injection surface as the skill processes user-supplied transcripts.
- Ingestion points: Video transcripts and configuration files (SKILL.md).
- Boundary markers: Absent from the provided instructional snippets.
- Capability inventory: Ability to execute shell commands and scripts via subprocesses.
- Sanitization: Input validation is not featured in the example code.
Audit Metadata