ai-video
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No executable code, scripts, or binaries are included in this skill. It is composed entirely of Markdown documentation and JSON evaluation sets.
- [SAFE]: There are no patterns of unauthorized data access, credential harvesting, or external network communication to untrusted domains.
- [INDIRECT_PROMPT_INJECTION]: While the skill provides an attack surface by processing user-supplied video requirements and mentions external content (comments/DMs/web results), it mitigates risk through explicit instructions. 1. Ingestion points: User prompts and web-sourced content (SKILL.md). 2. Boundary markers: Utilizes the structured BRIEF framework (references/the-brief-framework.md). 3. Capability inventory: Limited to generating text-based briefs for other tools. 4. Sanitization: Includes a specific instruction to treat external content strictly as data rather than commands (SKILL.md).
Audit Metadata