Eckra

The AI-powered Git CLI that turns "working on stuff" into meaningful commits.

---

💡 Overview

eckra is an interactive Git management tool designed for developers who value both speed and clarity. It integrates with multiple AI providers to analyze your staged changes and suggest context-aware commit messages, ensuring your project history remains professional and descriptive without the manual overhead.

🚀 Key Features

• 🤖 AI-Powered Suggestions: Automatically generates commit messages based on actual code diffs. Supports LM Studio, OpenAI, Anthropic, Ollama, OpenRouter, and Google Gemini.
• 📝 Select & Edit: Pick an AI suggestion and refine it instantly to match your specific needs.
• 🔍 Staged Diff Review: Inspect your changes in a beautiful, syntax-highlighted format directly before committing.
• 🎯 Interactive Dashboard: A comprehensive menu system for staging files, managing branches, stashing changes, and syncing with remotes.
• ⚡ Built for Speed: Zero-config required for standard Git operations. Fast, responsive, and intuitive.

📦 Installation

Install eckra globally using npm:

npm install -g eckra

🛠 Usage

Just type eckra in any Git repository to launch the interactive dashboard:

eckra

Quick Commands

Skip the menu and jump straight into action:

Command	Action
eckra commit	Start the AI-assisted commit flow
eckra status	Check repository status and staged files
eckra push	Sync local commits with the remote repository
eckra branch	Open the interactive branch manager

⚙️ AI Configuration

eckra supports multiple AI providers. You can switch between them using the built-in settings menu (Settings > Change Provider).

Supported Providers

Provider	Type	Default Model
LM Studio	Local	— (user-configured)
Ollama	Local	llama3
OpenAI	Cloud (API Key)	gpt-4o
Anthropic	Cloud (API Key)	claude-3-5-sonnet-20240620
OpenRouter	Cloud (API Key)	openai/gpt-4o
Google Gemini	Cloud (API Key)	gemini-2.0-flash

Default Setup (LM Studio)

By default, eckra connects to LM Studio's local server:

• URL: http://localhost:1234
• Requirement: Ensure LM Studio is running and the "Local Server" is started with a loaded model.

Configuration

You can configure your provider in two ways:

1. Interactive: Run eckra, go to More > Settings, and select your provider and enter your credentials.
2. Config file: Edit ~/.eckra/config.json directly:

{
  "aiProvider": "openrouter",
  "openrouterApiKey": "sk-or-...",
  "openrouterModel": "anthropic/claude-3.5-sonnet"
}

You can also create a .eckrarc file in your project root to override global settings per-repository.

🤝 Contributing

Contributions make the open-source community an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

Please see CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.

📄 License

Distributed under the MIT License. See LICENSE for more information.