video-transitions
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute
ffmpeganduv run video-agentcommands for video processing, metadata extraction, and result verification. These operations are essential to the skill's stated purpose of video editing. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface where it reads and parses the output of
video-agent info(specifically video duration values). This output is then interpolated into subsequent shell commands. While this represents a theoretical injection surface if file metadata were maliciously crafted, the processing is restricted to numeric offset calculations for ffmpeg filters.
Audit Metadata