Command AI Assistant

A ZSH plugin that uses Ollama to analyze command execution and provide instant AI-powered feedback when commands fail or produce unexpected results.

Features

• 🔍 Automatically monitors command execution
• ❌ Detects failed commands and non-zero exit codes
• 🤖 Provides AI-powered analysis using Ollama
• ✨ Clean, colorized output in your terminal
• ⚡ Lightweight and fast (only analyzes when needed)

Prerequisites

• ZSH shell
• Ollama installed and running
• A compatible Ollama model (default: llama3.2)

Installation

1. Install Ollama:

   macOS/Linux:

   curl -fsSL https://ollama.ai/install.sh | sh

2. Start Ollama service

   ollama serve

3. Pull the default model:

   ollama pull llama3.2

4. Install the plugin:

   Create plugins directory if it doesn't exist

   mkdir -p ~/.zsh/plugins

   Clone this repository

   git clone https://github.com/yourusername/command-ai-assistant.git
   cp command-ai-assistant/last-command.plugin.zsh ~/.zsh/plugins/

5. Add to your ~/.zshrc:

   source ~/.zsh/plugins/last-command.plugin.zsh

6. Reload your shell:

   source ~/.zshrc

🚀 Usage

Simply use your terminal as normal. The plugin will automatically:

• Monitor command execution
• Show exit codes
• Provide AI analysis for failed commands

Example output:

$ ls /nonexistent
ls: cannot access '/nonexistent': No such file or directory
[14:23:47] ✗ (2) ls /nonexistent
💡 Analysis:
The command failed because the specified directory doesn't exist. Use 'ls -la' to check the current directory contents.
-------------------

Configuration

You can customize the plugin by editing last-command.plugin.zsh:

• Change the AI model:

In analyze_command function, change:

local analysis=$(ollama run llama2 "$prompt") # or any other model

• Adjust when analysis triggers:

In precmd function, modify conditions:

if [ $EXIT_CODE -ne 0 ] || [[ ${#LAST_COMMAND} -gt 10 ]]; then
# Adjust conditions here
fi

How It Works

1. Uses ZSH's preexec and precmd hooks to monitor commands
2. Captures command output and exit codes
3. Sends failed commands to Ollama for analysis
4. Displays formatted results in your terminal

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT License - feel free to use this in your own projects!

Acknowledgments

• Ollama for providing the AI backend
• ZSH community for the excellent shell hooks system

Support

If you encounter any issues or have questions:

1. Check the Issues page
2. Create a new issue if needed
3. Include your:
   • OS version
   • ZSH version
   • Ollama version
   • Error messages