dev-mate-cli

A command-line tool that leverages OpenAI's Chat Completion API to document code with the assistance of AI models.

Watch this Demo video to view features.

Features

• Source Code Documentation: Automatically generate comments and documentation for your source code.
• Multiple File Processing: Handle one or multiple files in a single command.
• Model Selection: Use AI model of your choice with the --model flag.
• Custom Output: Output the results to a file with the --output flag, or display them in the console.
• Stream Output: Stream the LLM response to command line with --stream flag.

Installation

npm install -g dev-mate-cli

Environment Variables

dev-mate-cli needs API_KEY and BASE_URL to generate responses, these variables should be stored in a .env file within the current directory. Make sure to use the API_KEY and BASE_URL from the same OpenAI-compatible completion API provider.

API_KEY=your_api_key
BASE_URL=https://api.openai.com/v1

Popular providers - OpenRouter, Groq, OpenAI.

Usage

Basic Usage

To run the tool, specify one or more source files or folders as input:

dev-mate-cli ./examples/file.js

For processing multiple files:

dev-mate-cli ./examples/file.js ./examples/file.cpp

For processing folders:

dev-mate-cli ./examples

Command-line Options

• -m, --model <model-name>: Choose the AI model to use (default: google/gemma-2-9b-it:free from OpenRouter).

  dev-mate-cli file.js -m "openai/gpt-4o-mini"

• -o, --output <output-file>: Write the output to a specified file.

  dev-mate-cli file.js -o output.js

• -t, --temperature <value>: Set the creativity level of the AI model (default: 0.7).

  dev-mate-cli file.js -t 1.1

• -u, --token-usage: Display token usage information

  dev-mate-cli file.js -u

• -s, --stream: Stream response to command line

  dev-mate-cli file.js -s

Additional Commands

• Check Version: To check the current version of the tool, use:

  dev-mate-cli --version

• Help: Display the help message listing all available options:

  dev-mate-cli --help

Examples

• Document a JavaScript file and save the result:

  dev-mate-cli ./examples/file.js --output file-documented.js --model google/gemini-flash-8b-1.5-exp

• Process multiple files and print output to the console:

  dev-mate-cli ./examples/file.js ./examples/file.py --model google/gemini-flash-8b-1.5-exp

LLM Configuration

To use a file for LLM configuration, create a dotfile named .dev-mate-cli.toml in the home directory of your system.

Ex: ~/.dev-mate-cli.toml:

model = "gpt-4o"
temperature = "1"

Contributing

Contributions are welcome! If you find a bug or have an idea for an improvement, feel free to open an issue or submit a pull request, view Contribution Guidelines for more details.

License

This project is licensed under the MIT License.