Skills
Navigate
skills/openbmb/minicpm/minicpm5-deploy-mlx/Gen Agent Trust Hub

minicpm5-deploy-mlx

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the mlx-lm and gguf Python packages from public registries to enable model execution.
  • [EXTERNAL_DOWNLOADS]: The skill downloads model weights from Hugging Face repositories based on user-provided variables (e.g., openbmb/MiniCPM5-1B-MLX).
  • [COMMAND_EXECUTION]: The skill executes mlx_lm.generate, mlx_lm.convert, and mlx_lm.server commands locally to manage the machine learning model workflow.
  • [COMMAND_EXECUTION]: The skill uses curl to interact with a local server hosted on 127.0.0.1 for testing the OpenAI-compatible API.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 07:34 AM
Security Audit — agent-trust-hub — minicpm5-deploy-mlx