video-editing
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides numerous templates and scripts for running FFmpeg and shell commands to cut, concatenate, and reframe video files locally as part of its primary workflow.
- [EXTERNAL_DOWNLOADS]: The skill facilitates interaction with well-known media generation platforms including ElevenLabs and fal.ai for creating audio and visual assets.
- [PROMPT_INJECTION]: The skill's workflow involves analyzing external transcripts to generate edit decisions, which presents a surface for indirect prompt injection.
- Ingestion points: External transcripts provided for analysis in Layer 2 (SKILL.md).
- Boundary markers: The provided prompt template lacks explicit delimiters for the transcript data.
- Capability inventory: The agent is instructed to generate shell scripts and FFmpeg commands (SKILL.md).
- Sanitization: There is no mention of validating or escaping transcript content before it is used to generate logic.
Audit Metadata