env-doctor

概述

Env-Doctor 是一个全面的工具，用于诊断和修复 GPU 环境兼容性问题，特别关注 NVIDIA 驱动程序、系统工具包、cuDNN 和 Python AI 库之间的 CUDA 版本不匹配问题。MCP 服务器向 Claude Desktop 等 AI 助手暴露 11 个诊断工具，使它们能够检查 GPU 环境、验证 Dockerfile、确定模型兼容性并提供修复方案。它在本地系统、Docker 容器和 CI/CD 流水线中均可使用，支持 WSL2 和多个云平台。

试试问 AI

装完之后，这里有 5 个你可以让 AI 做的事：

• check

  Perform full environment compatibility diagnosis between GPU driver, CUDA toolkit, cuDNN, and Python libraries.

  试试:Check my GPU environment for compatibility issues试试:Diagnose why PyTorch is failing with CUDA errors
• python-compat

  Check Python version compatibility with AI libraries and detect dependency cascade issues.

  试试:Check if Python 3.13 is compatible with my installed libraries试试:Show Python version conflicts with TensorFlow and PyTorch
• cuda-install

  Install CUDA Toolkit with automatic platform detection and verification.

  试试:Install CUDA toolkit for my system试试:Show me what CUDA installation commands would run
• install

  Get safe installation commands for AI libraries that match your CUDA version.

  试试:Get the correct PyTorch installation command for my system试试:Install flash-attn with proper CUDA compatibility
• model

  Check if AI models fit in GPU memory and get cloud GPU recommendations.

  试试:Check if Llama 3 70B will fit on my GPU试试:Show cloud GPU recommendations for a model
• dockerfile

  Validate Dockerfiles for GPU configuration and Python library compatibility.

  试试:Validate this Dockerfile for GPU issues试试:Check if my Dockerfile has the right CUDA image
• docker-compose

  Validate docker-compose.yml files for GPU configuration issues.

  试试:Check my docker-compose.yml for GPU configuration problems试试:Validate docker-compose setup for PyTorch container
• cuda-info

  Provide detailed analysis of CUDA toolkit installation and configuration.

  试试:Show detailed information about my CUDA installation试试:Detect multiple CUDA installations on my system
• cudnn-info

  Analyze cuDNN library installation and compatibility with CUDA.

  试试:Check if cuDNN is properly installed试试:Show cuDNN version compatibility with my CUDA
• scan

  Scan code for deprecated CUDA-related imports.

  试试:Scan my code for deprecated PyTorch imports试试:Find CUDA API calls that need updating
• debug

  Provide verbose diagnostic output for troubleshooting environment issues.

  试试:Show detailed debug information for my GPU environment试试:Get verbose output for CUDA configuration problems
• init

  Generate CI/CD workflow files for GPU environment monitoring.

  试试:Create a GitHub Actions workflow for GPU environment checks试试:Generate CI configuration for monitoring CUDA compatibility

可对比工具

安装

pip install env-doctor

{
  "mcpServers": {
    "env-doctor": {
      "command": "env-doctor-mcp"
    }
  }
}

FAQ

Env-Doctor 解决什么问题？

它诊断并修复 GPU/CUDA 兼容性问题，特别是 NVIDIA 驱动程序、CUDA 工具包、cuDNN 和 PyTorch、TensorFlow 等 Python AI 库之间的版本不匹配问题。

MCP 服务器如何与 AI 助手集成？

它通过模型上下文协议向 AI 助手暴露 11 个诊断工具，使它们能够检查 GPU 环境、验证配置并建议修复方案。

env-doctor 对比