video
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill follows best practices for guiding users through video production tasks and provides instructions for using established, reputable industry tools.
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill recommends installing
hyperframesvia npm and usingcreate-videovia npx. These are official tools for the HeyGen and Remotion platforms, which are well-known services in the video production domain. The references are used for their intended primary purpose and are documented neutrally. - [INDIRECT_PROMPT_INJECTION]: The skill includes a workflow for reverse-engineering video edits that involves ingesting untrusted data from external sources via tools like
watch-videoandsocial-fetch(located inreferences/edit-anatomy.md). This ingestion is a standard surface for the skill's primary function of style analysis and is considered safe in this context, especially with the included manual review step for the generated beat sheets.
Audit Metadata