voice-realism
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by design.
- Ingestion points: The skill parses user-supplied video prompts and dialogue scripts as the primary input for its rewriting workflow in SKILL.md.
- Boundary markers: There are no instructions to use delimiters or explicit warnings to ignore instructions embedded within the processed user data.
- Capability inventory: The skill uses file manipulation tools (Read, Write, Edit) and references execution scripts (gen_video.py, render_video.py) for final media production.
- Sanitization: The instructions do not include methods for sanitizing or escaping the user-provided content before interpolating it into the rewritten output.
Audit Metadata