pls

A command-line tool that uses AI to generate and execute terminal commands from natural language descriptions.

Overview

pls is a macOS terminal assistant powered by local AI (via Ollama) that translates your plain English requests into executable shell commands. It can also answer technical questions about macOS in a conversational mode.

Features

• 🤖 Natural Language to Commands: Describe what you want to do, get the exact command
• 💬 Ask Mode: Get technical answers and explanations with the -a flag
• ⚡ Smart Ollama Management: Automatically starts/stops Ollama server as needed
• 🎨 Beautiful Terminal UI: Rich formatting with syntax highlighting and interactive prompts
• 🔒 Safety First: Always shows the command and asks for confirmation before execution

Requirements

• Python 3.x
• macOS
• Ollama installed
• llama3.1:8b model pulled (ollama pull llama3.1:8b)

Python Dependencies

pip install ollama rich requests

Installation

1. Clone or download the script
2. Make it executable:

   chmod +x pls

3. (Optional) Move to your PATH:

   sudo mv pls /usr/local/bin/

Usage

Command Mode (Default)

Generate and execute commands from natural language:

pls "find all python files in current directory"
pls "compress Documents folder"
pls "show disk usage"

The tool will:

1. Generate the appropriate command
2. Display it with syntax highlighting
3. Ask for confirmation before running

Ask Mode (-a)

Get answers to technical questions:

pls -a "how do I change file permissions?"
pls -a "what's the difference between chmod and chown?"
pls -a "explain the find command"

How It Works

1. Ollama Detection: Checks if Ollama server is running
2. Auto-Start: Starts Ollama automatically if needed (with a neat loading animation)
3. AI Processing: Sends your query to the llama3.1:8b model
4. Smart Output:
   • Command mode: Shows executable command with confirmation prompt
   • Ask mode: Displays helpful answer in a formatted panel
5. Cleanup: Shuts down Ollama if it wasn't running before (saves resources)

Configuration

Edit these variables at the top of the script to customize:

MODEL = "llama3.1:8b"              # Change AI model
OLLAMA_API_URL = "http://127.0.0.1:11434"  # Change Ollama endpoint

Examples

# Find large files
pls "find files larger than 100MB"

# Git operations
pls "create a new git branch called feature-x"

# System info
pls "show CPU temperature"

# Ask for help
pls -a "how do I monitor network traffic?"

Safety Features

• Commands are always shown before execution
• User confirmation required (Y/n prompt)
• Graceful interrupt handling (Ctrl+C)
• Automatic cleanup of zombie processes

Troubleshooting

Ollama won't start:

• Ensure Ollama is installed: brew install ollama
• Check if the model is available: ollama list

Model not found:

ollama pull llama3.1:8b

Permission errors:

• Make sure the script is executable: chmod +x pls

License

This is a utility script provided as-is. Modify and use freely under the MIT license.

Contributing

Feel free to fork and customize for your needs! Some ideas:

• Add support for other AI models
• Implement command history
• Add shell completion
• Support for other operating systems

---

Note: This tool executes commands on your system. Always review generated commands before confirming execution.