claude-video-transcode

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill employs the Bash tool to execute ffmpeg, ffprobe, and several internal helper scripts (preflight.sh, detect_gpu.sh, estimate_size.sh) for processing video files and detecting hardware capabilities.
  • [SAFE]: The shell command templates correctly use double quotes around variables (e.g., "$INPUT", "$OUTPUT"), which is a recommended security practice to prevent command injection from file paths containing spaces or shell metacharacters.
  • [SAFE]: The skill performs metadata analysis using ffprobe with structured JSON output, providing a safe method for the agent to inspect input file properties without executing untrusted content.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 02:55 PM
Security Audit — agent-trust-hub — claude-video-transcode