ribbon-wrap
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local script located at
~/.claude/skills/video-effects/run.cjsto process video files and apply effects using Node.js. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided text via the
--wordsargument, which constitutes an attack surface for indirect prompt injection if the rendering engine does not handle special characters or instructions correctly. - Ingestion points: User-supplied text via the
--wordscommand-line argument. - Boundary markers: None explicitly mentioned in the instructions.
- Capability inventory: The underlying engine (
video-effects) performs file system operations (reading/writing MP4s) and utilizes Node.js, ffmpeg, and Puppeteer. - Sanitization: No sanitization logic is described in the skill file.
Audit Metadata