cutagent-clipping
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external media content such as videos and transcripts which represent ingestion points for untrusted data.
- Ingestion points: The skill analyzes long videos, podcasts, and interviews.
- Boundary markers: There are no instructions to use delimiters or to ignore commands that may be embedded in transcripts.
- Capability inventory: The skill has the ability to create and modify DaVinci Resolve timelines using the CutAgent CLI.
- Sanitization: There is no mention of sanitizing content extracted from media before it is used for timeline naming or reporting.
Audit Metadata