brand-video-generator
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains only process instructions for an AI agent to follow when creating brand videos. It does not include scripts, shell commands, or external dependencies. References to internal skills like 'Editframe composition' are standard for multi-skill agent environments.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external data from brand websites and user assets, which is a potential surface for indirect prompt injection.
- Ingestion points: External brand pages and user-supplied assets (SKILL.md).
- Boundary markers: Absent.
- Capability inventory: Video rendering and composition via internal Editframe skills.
- Sanitization: Instructions include steps to verify facts and distinguish brand claims from interpretations, which helps mitigate risks.
Audit Metadata