Chatter

A command-line interface application for interacting with OpenAI's GPT models using the DecaChat wrapper.

Features

• 🖥️ Interactive command-line interface
• 💬 Real-time chat with AI assistant
• 📜 Conversation history viewing
• 🔄 Conversation reset capability
• ⚙️ Configurable AI parameters
• 🔐 Secure environment variable configuration

Prerequisites

• Node.js (v14 or higher)
• npm (Node Package Manager)

Installation

1. Clone the repository:

git clone <repository-url>
cd decachat-cli

2. Install dependencies:

npm install

3. Create a .env file in the project root with your configuration:

API_KEY=your-openai-api-key
BASE_URL=https://api.openai.com/v1
CHAT_MODEL=gpt-4o-mini

Usage

1. Start the application:

node app.js

2. Available commands:

• Type your message and press Enter to chat with the assistant
• !clear - Reset the conversation
• !history - View conversation history
• quit - Exit the application

Configuration

The application can be configured using the following environment variables:

Variable	Description	Default Value
API_KEY	Your OpenAI API key	Required
BASE_URL	API endpoint URL	https://api.openai.com/v1
CHAT_MODEL	GPT model to use	gpt-4o-mini

Additional configuration in the code:

• maxTokens: 150 (maximum response length)
• temperature: 0.7 (response creativity level)

Example Session

DecaChat Test Application
------------------------
Type your messages and press Enter. Type "quit" to exit.
Special commands:
  !clear - Clear conversation history
  !history - Show conversation history
------------------------

You: Hello, who are you?
Assistant is typing...
Assistant: I'm an AI assistant programmed to provide concise and helpful responses.

You: !history
Conversation History:
[system]: You are a helpful assistant who provides concise responses.
[user]: Hello, who are you?
[assistant]: I'm an AI assistant programmed to provide concise and helpful responses.

You: !clear
Conversation cleared.

You: quit

Error Handling

The application includes robust error handling for:

• API connection issues
• Invalid responses
• Token limit exceeded
• Network errors

Errors are logged to the console with descriptive messages.

Development

The application is built using:

• deca-chat: Wrapper for OpenAI's API
• dotenv: Environment variable management
• readline: Command-line interface handling

To modify the system message or other parameters, edit the values in the main function:

chat.setSystemMessage('You are a helpful assistant who provides concise responses.');

Contributing

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

License

MIT License - see the LICENSE file for details.