McpEnvInstaller

English | 简体中文

McpEnvInstaller is a cross-platform script toolset for automating the deployment of the MCP (Model Context Protocol) runtime environment.

MCP is an open protocol that standardizes how applications provide context to LLMs. Think of MCP like a USB-C port for AI applications. Just as USB-C provides a standardized way to connect your devices to various peripherals and accessories, MCP provides a standardized way to connect AI models to different data sources and tools. For more information about MCP, please visit https://modelcontextprotocol.io/introduction

✨ Project Description

This project is dedicated to simplifying and automating the complex process of setting up the MCP runtime environment across different operating systems.

The MCP STDIO protocol allows interaction with local programs via standard input and output, and typically relies on two command-line tools:

• uvx: Used for quickly executing command-line tools or entry points provided by Python packages installed via uv or pip, often in isolated environments.
• npx: Used for easily running command-line tools installed via the Node.js package manager (npm), without requiring global installation.

The proper functioning of these executors (uvx and npx) depends on their respective runtime foundations: the Python tools executed by uvx require a Python runtime, while npx and the Node.js tools it executes require a Node.js runtime

Therefore, the core objective of this project is to detect, install, and configure the following key dependencies required for the MCP environment:

• Python: Provides the necessary Python runtime environment.
• uv: Serves as a high-performance Python package manager and provides the uvx executor.
• Node.js: Provides the Node.js runtime environment and includes the npx tool.

To ensure that most MCP projects and other libraries they depend on (such as crawl4ai) can function correctly, using Python 3.10 or higher is strongly recommended. Please note that while uv itself does not run on Python, adhering to this recommendation is advised for compatibility with commonly used libraries in the ecosystem and for a better development experience.

Similarly, to ensure good compatibility with tools and applications within the MCP ecosystem, we also recommend using Node.js LTS 16.0 or higher.

By running the corresponding platform scripts, these dependencies can be automatically detected and installed, allowing for rapid construction of the MCP runtime environment.

Supported Operating Systems and Architectures

Operating System	Architecture	Support Status	Test Environment
macOS	AMD64 (Intel)	✅ Supported / Tested	GitHub Workflow
macOS	ARM64 (Apple Silicon)	✅ Supported / Tested	macOS M1 15.4.1, GitHub Workflow
Windows	AMD64	✅ Supported / Tested	Windows 2019, Windows 11, GitHub Workflow: 2019, 2022
Windows	ARM64	✅ Supported / Tested	Windows 11
Linux (Red Hat-based)	AMD64 (x86_64)	✅ Supported / Tested	GitHub Workflow Docker: Rockylinux 9
Linux (Red Hat-based)	ARM64 (aarch64)	✅ Supported / Tested	Fedora 38, Openeuler25
Linux (Debian-based)	AMD64 (x86_64)	✅ Supported / Tested	Ubuntu 24, GitHub Workflow: Ubuntu 24, GitHub Workflow Docker: debian-latest
Linux (Debian-based)	ARM64 (aarch64)	✅ Supported / Tested	Ubuntu 22
Linux (Alpine, Opensuse, Archlinux)	AMD64 (x86_64)	✅ Supported / Tested	GitHub Workflow Docker
Linux (Alpine)	ARM64 (aarch64)	✅ Supported / Tested	Docker

• Linux: Compatible with major distributions (e.g., Debian/Ubuntu - apt, RHEL/CentOS/Fedora - yum/dnf, Alpine - apk, Arch - pacman, OpenSUSE - zypper, etc.), handled automatically by the linux.sh script.
• Windows: Supported via a PowerShell script (windows.ps1).
• macOS: Supported via a Shell script (macos.sh).

🚀 Usage

Preparation

1. Network Connection: The script needs to download software packages and dependencies (Python, Node.js, uv, etc.) from the internet during execution. Ensure you have a stable network connection.
2. Permissions:
   • Linux: sudo privileges are usually required to install system-level packages.
   • Windows: Running PowerShell with Administrator privileges is required to ensure software is installed correctly to system paths and environment variables are configured.
   • macOS: You may be asked to enter your user password during script execution to allow sudo commands (e.g., when installing dependencies using a system package manager).
3. Package Managers (macOS & Linux): The script will attempt to use the system's native or recommended package manager (e.g., apt, yum, dnf, pacman, zypper, apk, brew, port). If macOS users do not have Homebrew or MacPorts installed, the script will prompt and prioritize installing Homebrew.

Script Download Note

We provide download links from both GitHub and jsDelivr. If you experience network issues accessing GitHub directly (raw.githubusercontent.com), please try using the jsDelivr link first, as it offers better access speed and stability globally.

🍎 macOS

Key Features:

• Supports AMD64 and ARM64 (M1/M2) architectures.
• Intelligently detects and installs suitable versions of Python (>=3.10) and Node.js (>=16.0).
• Automatically selects Homebrew (preferred) or MacPorts for installation. If neither is present, it guides the installation of Homebrew.
• Installs uv and provides clear guidance for PATH configuration if needed.
• Automatically handles installation path differences for M1/M2 (ARM64) architecture.

Installation Steps:

Open Terminal and run one of the following commands:

• Recommended (jsDelivr):

  curl -fsSL https://cdn.jsdelivr.net/gh/aidyou/McpEnvInstaller@main/macos.sh | sh

• Alternate (GitHub Raw):

  curl -fsSL https://raw.githubusercontent.com/aidyou/McpEnvInstaller/main/macos.sh | sh

• Recommended (For chinese users):

  curl -fsSL https://gitee.com/aidyou/McpEnvInstaller/raw/main/macos.sh | sh

Important Notes:

• Network Proxy: If downloads are slow or fail, try setting up a network proxy:

  export http_proxy="http://your-proxy-server:port"
  export https_proxy="http://your-proxy-server:port"
  # Then run the curl command again

• Older macOS Versions: For older systems (e.g., macOS Mojave 10.14), you may need to manually install Xcode Command Line Tools (xcode-select --install) or use official Python installers. The script attempts compatibility but cannot guarantee perfect operation on all older versions.

• PATH Configuration: If uv or other tools are installed but you get command not found, please add paths like $HOME/.cargo/bin or /opt/homebrew/bin to your Shell configuration file (~/.zshrc (default for macOS Catalina+), ~/.bash_profile, ~/.bashrc, etc.) according to the script's output. For example, for Zsh users:

  echo 'export PATH="$HOME/.cargo/bin:$PATH"' >> ~/.zshrc
  source ~/.zshrc # Apply the configuration

  For Bash users:

  echo 'export PATH="$HOME/.cargo/bin:$PATH"' >> ~/.bash_profile
  source ~/.bash_profile # Apply the configuration

🪟 Windows

Key Features:

• Supports AMD64 (x64), ARM64, x86 architectures.
• Compatible with Windows 10 (1809+) and Windows 11.
• Automatically downloads and silently installs recommended versions of Python (>=3.10) and Node.js LTS (>=16.0).
• Automatically installs uv.

Installation Steps:

Right-click the "Start" button, select "Windows PowerShell (Admin)", and run:

• Recommended (jsdelivr):

irm https://cdn.jsdelivr.net/gh/aidyou/McpEnvInstaller@main/windows.ps1 | iex

• Alternate (GitHub Raw):

irm https://raw.githubusercontent.com/aidyou/McpEnvInstaller/main/windows.ps1 | iex

• Recommended (For chinese users):

irm https://gitee.com/aidyou/McpEnvInstaller/raw/main/windows.ps1 | iex

Important Notes:

• Execution Policy: If you encounter an error preventing script execution, you can temporarily relax the policy for the current process:

  Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass -Force
  # Then run the irm command again

• Installation Paths: Python is typically in %ProgramFiles%\PythonXX or %LOCALAPPDATA%\Programs\Python (depending on installation options), Node.js in C:\Program Files\nodejs.

• Verification: After installation, open a new terminal window and run python --version, node --version, uv --version to verify successful installation.

🐧 Linux

Key Features:

• Automatically detects the distribution and uses the corresponding package manager (apt, dnf, yum, pacman, zypper, apk, etc.).
• Attempts to install Python (>=3.10) and Node.js (>=16.0) meeting the version requirements.
• If the system repository version is too old, it will attempt to add reliable third-party sources (like deadsnakes PPA for Ubuntu, NodeSource) or prompt the user for manual action.
• Installs uv, prioritizing the official static binary.

Installation Steps:

Open a terminal and run one of the following commands:

• Recommended (jsDelivr):

  curl -fsSL https://cdn.jsdelivr.net/gh/aidyou/McpEnvInstaller@main/linux.sh | sh

• Alternate (GitHub Raw):

  curl -fsSL https://raw.githubusercontent.com/aidyou/McpEnvInstaller/main/linux.sh | sh

• Recommended (For chinese users):

  curl -fsSL https://gitee.com/aidyou/McpEnvInstaller/raw/main/linux.sh | sh

Important Notes:

• Network Proxy: Similar to macOS, if you have download issues, set http_proxy and https_proxy environment variables first.

• Older Distributions: For distributions that are no longer maintained (like CentOS 7), system packages might be very old. The script attempts compatibility, but you might need to perform additional steps (like enabling SCL, etc.) based on the prompts.

• Non-Root Users: The script defaults to using sudo for installation commands. Ensure your current user has sudo privileges.

• Silent Installation: You can add the -s -- -q parameters to reduce script output:

  curl -fsSL ... | sh -s -- -q

• Package Manager Differences: Package names and available versions may differ across distributions. The script attempts to adapt, but if you encounter issues on a specific distribution, feel free to open an Issue.

🤝 Contributing

Contributions of any kind are welcome! If you find a bug, the script doesn't work correctly on a platform, or you have suggestions for improvement, please feel free to create Issues or submit Pull Requests.

📜 License

This project is licensed under the MIT License.

The MIT License is a very permissive free software license. See the LICENSE file in the repository root for details.