media-transcoding

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions rely on the execution of local scripts (ffmpeg_convert.sh, convert_video.py) and the ffmpeg binary to perform media processing tasks.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of media filenames (e.g., "My Video.mp4"). If the referenced scripts do not properly sanitize these inputs before passing them to the shell, it could lead to command injection vulnerabilities.
  • Ingestion points: Input media filenames and metadata.
  • Boundary markers: The skill documentation demonstrates using quotes for filenames in examples.
  • Capability inventory: Local file reading, writing to converted/ and backup/ directories, and execution of FFmpeg commands.
  • Sanitization: The skill instructions recommend standard quoting, but full sanitization depends on the implementation of the external scripts referenced.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 12:53 PM
Security Audit — agent-trust-hub — media-transcoding