digital-human
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-supplied data, such as narration scripts and avatar names, which are then passed to backend API tools for video generation.
- Ingestion points: User-provided scripts, metadata names, and imported media files specified in the Required Inputs and Resolve a script sections of SKILL.md.
- Boundary markers: The instructions do not specify the use of delimiters or escaping mechanisms when interpolating user-generated scripts into tool calls for submission.
- Capability inventory: The skill uses manage_avatar and submit_avatar_video tools to interact with backend services for content creation.
- Sanitization: There are no explicit instructions for sanitizing or validating the content of user scripts to prevent injection attacks, relying instead on backend implementation.
Audit Metadata