video-ad-specs

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the belt CLI tool (associated with inference.sh) to execute various AI models for video generation (bytedance/seedance-2-0, google/veo-3-1-fast), text-to-speech (falai/dia-tts), and media processing (infsh/video-audio-merger, infsh/caption-videos). The allowed-tools configuration appropriately limits command execution to this specific tool set.
  • [EXTERNAL_DOWNLOADS]: The instructions reference external installation guides and skill packages from the inference-sh GitHub organization (github.com/inference-sh/skills). This is standard behavior for integrating with the described AI inference platform.
  • [DATA_EXFILTRATION]: Includes the belt login command, which is a legitimate authentication step for the service described and does not represent an exfiltration risk in this context.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 05:02 AM
Security Audit — agent-trust-hub — video-ad-specs