🌟 Lumina AI Terminal Agent v2.0

An advanced AI terminal agent powered by Groq with full system capabilities - similar to Claude CLI, but with MUCH MORE POWER!

✨ Features

🤖 Core Features

• Advanced Conversational AI - Natural and intelligent chat
• Bash Execution - Execute any system command automatically
• Real-time Streaming - Instant responses with live output
• Interactive Setup - API key saved automatically

🎨 New in v2.0

• 🔒 Safety First - Confirms dangerous commands before execution
• 🤖 Multiple AI Models - Switch between Llama, Mixtral, Gemma
• ⚙️ Work Modes - Developer, SysAdmin, DevOps, Security, Data Analysis
• 💾 Persistent History - Save and restore conversation sessions
• 📝 Snippets & Templates - Save frequently used commands
• 🔌 Plugin System - Extend functionality with custom plugins
• 📊 Data Visualization - Tables, charts, progress bars
• 🎨 Syntax Highlighting - Colorized code in responses
• 📦 Built-in Tools - File system, Git, packages, network, text processing

📦 Installation

npm install -g lumina-ai-agent

🚀 Usage

lumina

On first run, enter your Groq API key from: https://console.groq.com

💬 Commands

Basic Commands

• clear - Clear conversation history
• reset-key - Reset API key
• exit / quit - Exit program
• help - Show all commands

Advanced Commands

• model - View/change AI model
• model llama-3.1-8b - Switch to faster model
• mode - View/change work mode
• mode dev - Switch to developer mode
• history - Show saved sessions
• snippet list - List saved snippets
• snippet save <name> <command> - Save a snippet
• plugins - List loaded plugins

🤖 Available Models

• llama-3.3-70b - Most capable (default)
• llama-3.1-70b - Balanced performance
• llama-3.1-8b - Fastest
• mixtral-8x7b - Great for coding
• gemma-7b - Lightweight

⚙️ Work Modes

• default - General purpose
• dev - Software development (best practices, testing, debugging)
• sysadmin - System administration (monitoring, security)
• devops - CI/CD, Docker, automation
• security - Security analysis and auditing
• data - Data analysis and processing

🛠️ Available Tools

📁 File System

ls -lah                    # List files
cat file.txt               # Read file
echo "text" > file.txt     # Create file
mkdir folder               # Create directory
rm file.txt                # Delete file
cp source dest             # Copy file
mv source dest             # Move/rename
chmod +x script.sh         # Change permissions
find . -name "*.js"        # Search files

🔧 System & Processes

ps aux                     # List processes
top                        # Process monitor
kill PID                   # Kill process
free -h                    # Memory usage
df -h                      # Disk usage
uname -a                   # System info

📦 Git

git status                 # Repository status
git log                    # Commit history
git diff                   # View changes
git add .                  # Add files
git commit -m "msg"        # Commit
git push                   # Push to remote

📦 Package Managers

npm install package        # Install npm package
pip install package        # Install Python package
apt-get install package    # Install on Ubuntu/Debian
brew install package       # Install on macOS

🌐 Network

ping google.com            # Test connection
curl https://api.com       # Make HTTP request
wget url                   # Download file
netstat -tuln              # Open ports

📝 Text Processing

grep "pattern" file        # Search text
sed 's/old/new/' file      # Replace text
awk '{print $1}' file      # Process columns
head -n 10 file            # First lines
tail -n 10 file            # Last lines

🎯 Usage Examples

❯ List files in current directory
❯ What's the system memory usage?
❯ Show me git status
❯ Install express package via npm
❯ Search for "TODO" in all .js files
❯ What's my public IP?
❯ Create a test.txt file with "Hello World"
❯ Make a commit with message "fix: bug fixed"

🔌 Plugin System

Create custom plugins in ~/.lumina/plugins/

Example plugin:

export default {
  name: 'my-plugin',
  version: '1.0.0',
  description: 'My custom plugin',
  
  async init(pluginManager) {
    pluginManager.registerHook('beforeCommand', async (command) => {
      console.log('Executing:', command);
      return command;
    });
  }
};

📝 Snippets

Built-in snippets:

• git-quick - Quick git add, commit, push
• npm-init - Initialize npm project
• docker-clean - Clean Docker resources
• find-large - Find large files
• port-check - Check port usage

Save custom snippets:

❯ snippet save deploy "git push && npm run build"

🔒 Safety Features

Lumina will ask for confirmation before executing:

• rm -rf commands
• chmod 777 operations
• System shutdown/reboot
• Disk formatting
• Other potentially dangerous operations

📁 Project Structure

lumina/
├── src/
│   ├── core/
│   │   ├── config.js          # Models & configuration
│   │   ├── configManager.js   # API key manager
│   │   ├── conversation.js    # Chat manager
│   │   ├── processor.js       # Response processor
│   │   ├── history.js         # Session history
│   │   ├── snippets.js        # Snippet manager
│   │   ├── plugins.js         # Plugin system
│   │   └── workModes.js       # Work modes
│   ├── tools/
│   │   ├── bash.js            # Bash executor
│   │   ├── filesystem.js      # File operations
│   │   ├── system.js          # System monitoring
│   │   ├── git.js             # Git operations
│   │   ├── package.js         # Package managers
│   │   ├── network.js         # Network operations
│   │   ├── text.js            # Text processing
│   │   └── manager.js         # Tool manager
│   ├── utils/
│   │   ├── formatter.js       # Colors & formatting
│   │   ├── safety.js          # Command safety
│   │   ├── syntax.js          # Syntax highlighting
│   │   └── visualization.js   # Charts & tables
│   └── index.js               # Main CLI
└── package.json

🔧 Configuration

All configuration is stored in ~/.lumina/:

• config.json - API key
• history/ - Conversation sessions
• snippets/ - Saved snippets
• plugins/ - Custom plugins

🛡️ Security

• API key stored locally in your home directory
• Dangerous commands require confirmation
• 30s timeout for long commands
• Limited buffer for large outputs (10MB)
• Errors are captured and reported
• System permissions are respected

🚀 Advanced Capabilities

• Command chaining - Multiple commands with && and |
• Persistent context - Maintains conversation history
• Intelligent analysis - Automatically decides when to execute commands
• Real-time feedback - Response streaming
• Error handling - Captures and explains errors
• Multi-platform - Works on Linux, macOS and Windows (WSL)
• Extensible - Plugin system for custom functionality

📝 License

MIT

Made with ❤️ using Groq AI

NPM Package: https://www.npmjs.com/package/lumina-ai-agent