tencent-hunyuanvideo

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill implements a 'dry-run-first' architecture that mandates explicit user approval and monetary caps before any paid API calls or large-scale model downloads are initiated.
  • [EXTERNAL_DOWNLOADS]: Fetches necessary model weights, configuration files, and official source code from well-known and reputable repositories including GitHub (Tencent-Hunyuan), Hugging Face, and ModelScope.
  • [COMMAND_EXECUTION]: Utilizes system commands for essential video processing tasks (using ffmpeg and ffprobe) and model operations (using git, pip, and torchrun) which are consistent with the skill's documented primary purpose.
  • [SAFE]: Includes sophisticated validation mechanisms, such as verifying host suffixes for result URLs, checking for public-only IP addresses during downloads, and performing SHA-256 integrity checks on artifacts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 02:28 PM
Security Audit — agent-trust-hub — tencent-hunyuanvideo