Skills
Navigate
skills/mini-yifan/open-source-video-gen-skill/minimax-h3/Gen Agent Trust Hub

minimax-h3

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The scripts/connect_server.py script executes shell commands using subprocess.run() to manage SSH connections via expect or sshpass for remote server discovery.
  • [COMMAND_EXECUTION]: The scripts/prepare_assets.sh script invokes ffmpeg through shell execution to generate placeholder assets.
  • [COMMAND_EXECUTION]: Markdown instructions in SKILL.md direct the agent to source shell configuration files (e.g., source ~/.zshrc), which can result in the execution of arbitrary user-defined shell code.
  • [DATA_EXPOSURE]: The skill reads sensitive local configuration files such as ~/.config/autodl.env to retrieve authentication tokens and passwords required for server interaction.
  • [EXTERNAL_DOWNLOADS]: The scripts/poll_video.py script downloads generated MP4 files from remote endpoints specified by the SEETACLOUD_BASE_URL environment variable.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 09:40 AM