video-prompting
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists exclusively of Markdown and CSV files containing instructions and examples. There are no scripts (Python, JavaScript, shell, etc.) or executable files included in the package.
- [SAFE]: Analysis of the instructions across all 20 files shows they are limited to directing the AI to act as a prompt engineering specialist. The skill focuses on routing user requests to the correct model guidance (e.g., MiniMax H3, Ovi, LTX-2.5) and following specific formatting rules for video generation prompts.
- [SAFE]: All external URLs mentioned in the files point to official documentation, official blog posts, or official code repositories of well-known and trusted AI organizations (e.g., Hugging Face, Google Cloud, Lightricks/LTX, ByteDance, Character.AI, GitHub). These references are used solely for informational context.
- [SAFE]: The skill strictly follows best practices for privacy and security within prompts, repeatedly instructing the AI to exclude API parameters, model versions, and sensitive settings from the final output unless explicitly requested by the user outside of the prompt itself.
Audit Metadata