MXSH

A universal shell that eliminates operating system barriers through intelligent command translation and AI-powered security validation.

Overview

MXSH is a cross-platform command-line shell designed to provide a unified interface across Windows, Linux, and macOS. Instead of requiring users to learn platform-specific command syntax, MXSH automatically detects and converts commands to work on any system while maintaining security through comprehensive validation.

Key Features

Universal Command Translation Execute commands in any shell syntax. MXSH automatically converts platform-specific commands to work on your current system. Use apt on Windows, winget on Linux, or brew anywhere.

AI-Powered Security Analysis Every command undergoes intelligent risk assessment before execution. The system evaluates potential dangers, destructive operations, and security implications through OpenAI's API.

DNS-Based Download Protection Network downloads via curl or wget are validated against Cloudflare's malware-blocking DNS resolver, preventing connections to known malicious domains.

Encrypted Environment Management Store sensitive environment variables with AES-256-GCM encryption. Support for both machine-specific and password-based encryption ensures credential security.

Multi-Shell Support Execute commands in PowerShell, Bash, or CMD from a single interface using shell-specific prefixes or automatic routing.

Quick Start

Installation

Build from source using Rust:

git clone https://github.com/Mxshellit/Mxsh.git
cd Mxsh
cargo build --release

Initialize

Set up your OpenAI API key:

mxsh init

Use

Run commands naturally and let MXSH handle the rest:

apt install nodejs

On Windows, MXSH automatically converts this to winget install nodejs.

Documentation

Comprehensive documentation is available in the docs/ directory:

• Introduction - Overview and core concepts
• Getting Started - Installation and setup guide
• Features - Detailed feature documentation
• Commands - Complete command reference
• Security - Security model and best practices
• Configuration - Customization and settings

Requirements

• Operating System: Windows, Linux, or macOS
• Rust toolchain (for building from source)
• OpenAI API key (for AI-powered features)

Built-In Commands

• help - Display command reference
• init - Initialize MXSH configuration
• whoami - Show current user and OS
• models - Select AI model
• shells - Configure default shell
• cpu - Display top processes by CPU usage
• cd - Enhanced directory navigation
• env - Manage encrypted environment variables
• clear - Clear screen and redisplay banner

Shell Prefixes

Execute commands in specific shells:

$: Get-Process       # PowerShell
#: ps aux             # Bash
>: dir /w            # CMD

Security Features

Risk Levels Commands are classified as SAFE, LOW, MEDIUM, HIGH, or CRITICAL based on potential impact. High-risk operations require explicit user confirmation or bypass flags.

Whitelisted Commands Common read-only commands like ls, cat, and git execute immediately without AI analysis.

Bypass Flags For trusted operations:

• --skip-ai-check - Bypass AI security analysis
• --skip-dns-check - Skip DNS validation for downloads

Contributing

Contributions are welcome. Please refer to the repository's contribution guidelines for details on submitting issues, feature requests, and pull requests.

License

See the LICENSE file for details.

Support

For issues and questions, visit the GitHub repository.