B-roll Generation
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of scripts, transcripts, and Edit Decision Lists (EDLs) which could contain malicious instructions.
- Ingestion points: Untrusted content enters the agent context through user-provided script/transcript files and source edits as specified in the "Required inputs" section of SKILL.md.
- Boundary markers: The skill instructions do not define explicit boundary markers or provide the agent with directions to ignore potential commands embedded within the input data.
- Capability inventory: The skill utilizes the muapi capability set to generate video and images and to upload files (media.generate_video, media.generate_image, media.upload_file).
- Sanitization: There is no evidence of input sanitization, validation, or escaping of the ingested text content to mitigate risks of instruction injection.
Audit Metadata