lampcode

🚀 LampCode

A powerful CLI coding assistant powered by OpenRouter API. Get AI help with your coding tasks directly from the terminal.

✨ Features

• Interactive Chat: Natural language coding assistance
• Project Analysis: Understands your current project structure
• File Operations: Read, analyze, and help modify your code files
• Autonomous Tooling: AI can invoke built-in tools to read, search, apply precise edits, or create files on your behalf
• Guided CLI Experience: Structured dividers and detailed tool summaries keep every interaction easy to follow
• Multiple AI Models: Support for Claude, GPT, and other OpenRouter models
• Context Aware: Provides relevant suggestions based on your project

🛠️ Installation

1. Clone and Setup

# Navigate to your project directory
cd your-project

# Install dependencies
npm install

2. Configure API Key

• Copy the environment template:

  cp .env.example .env
  

• Get your OpenRouter API key from openrouter.ai

• Add your API key to .env:

  OPENROUTER_API_KEY=your_api_key_here
  OPENROUTER_MODEL=anthropic/claude-3.5-sonnet
  

3. Make it globally available

Option A: Global npm install (Recommended)

npm install -g .

Option B: Add to PATH manually

# On Windows
npm run build
# Add the bin directory to your PATH

# On macOS/Linux
chmod +x bin/lamp.js
sudo ln -s $(pwd)/bin/lamp.js /usr/local/bin/lamp

🚀 Usage

Start Interactive Session

Simply type lamp in your terminal:

lamp

Available Commands

• help - Show available commands
• clear - Clear the terminal screen
• read <filename> - Read and analyze a specific file
• edit <filename> - Edit a file with AI suggestions
• search <query> - Search codebase for specific text
• open <filename> - Open a file in your default editor
• exit or quit - Exit LampCode
• AI chats can also autonomously trigger internal tools (like reading files, searching code, or creating new files) when additional context is needed.

Example Interactions

lamp> How can I improve the performance of this React component?

lamp> Write a function to validate email addresses in TypeScript

lamp> read package.json

lamp> edit app.js
# AI will help you modify the file with your instructions

lamp> search axios
# Find all occurrences of "axios" in your codebase

lamp> What does this error mean: TypeError: Cannot read property 'map' of undefined?

lamp> Create a todo list component with React hooks

🔧 Configuration

Supported Models

You can change the AI model in your .env file:

# Claude models (recommended)
OPENROUTER_MODEL=anthropic/claude-3.5-sonnet
OPENROUTER_MODEL=anthropic/claude-3-opus
OPENROUTER_MODEL=anthropic/claude-3-haiku

# OpenAI models
OPENROUTER_MODEL=openai/gpt-4
OPENROUTER_MODEL=openai/gpt-3.5-turbo

# Other models available on OpenRouter

Project Context

LampCode automatically analyzes your project files to provide context-aware assistance:

• Reads common file types (.js, .ts, .py, .java, etc.)
• Understands your project structure
• Provides relevant suggestions based on your codebase
• Keeps the context fresh when the AI requests a rescan or creates new files (with logging so you can review every tool call)

📝 Requirements

• Node.js 16.0.0 or higher
• OpenRouter API key
• Internet connection for AI requests

🐛 Troubleshooting

Common Issues

• "command not found"

  • Make sure LampCode is installed globally or added to PATH
  • Try running npm link in the project directory

• "API key not found"

  • Ensure OPENROUTER_API_KEY is set in your .env file
  • Check that the .env file is in your project root

• "Rate limit exceeded"

  • OpenRouter has rate limits - wait a moment and try again
  • Consider upgrading your OpenRouter plan for higher limits

• Permission denied

  • On macOS/Linux, make sure the script is executable: chmod +x bin/lamp.js

🤝 Contributing

• Fork the repository
• Create your feature branch: git checkout -b feature/amazing-feature
• Commit your changes: git commit -m 'Add amazing feature'
• Push to the branch: git push origin feature/amazing-feature
• Open a Pull Request

📄 License

This project is licensed under the MIT License.

🙏 Acknowledgments

• Built with OpenRouter API
• Inspired by modern CLI coding assistants
• Thanks to the open source community for amazing tools like Commander.js and Inquirer.js