veo-prompting
Pass
Audited by Gen Agent Trust Hub on Jun 12, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides detailed shell command templates for using
ffmpegto trim, merge, and apply transitions to video files. This is a core part of the multi-clip production workflow described in the skill and relies on theffmpegutility being available in the environment. - [PROMPT_INJECTION]: The skill defines a structured ingestion surface where user requests are transformed into complex AI prompts.
- Ingestion points: User descriptions provided during video generation or edit requests (SKILL.md).
- Boundary markers: None explicitly defined for user input within the provided formula templates.
- Capability inventory: Execution of
veotools (veo_generate_video, veo_image_to_video, etc.) andffmpegshell commands. - Sanitization: No specific sanitization or escaping of user input is documented in the prompt engineering layers.
Audit Metadata