create-kinetic-typography

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands via ffmpeg for concatenating images and muxing video files. Specifically, it uses the -f concat and -safe 0 flags to process locally generated PNG sequences.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted text content from users to be rendered into kinetic typography. While this is a data ingestion surface, the rendering to a pixel-based format (PNG) serves as a functional boundary for the text data.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill accesses local system font paths such as /System/Library/Fonts/Supplemental/Impact.ttf to perform its primary function. No sensitive user data access or external network exfiltration was detected.
  • [DYNAMIC_EXECUTION]: The skill generates a temporary file list (list.txt) and a sequence of PNG images which are then executed upon by a shell-based tool (ffmpeg) to produce the final video artifact.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 04:39 PM
Security Audit — agent-trust-hub — create-kinetic-typography