🧠 gitcommit — AI Commit Message Generator

Generate meaningful, conventional git commit messages using AI — from your terminal.

✨ Features

• 🤖 Multi-provider AI — OpenAI, Anthropic, Google Gemini, Groq, NVIDIA NIM, OpenRouter, Ollama
• 📝 Conventional Commits — Automatically follows the spec (feat, fix, chore, docs, etc.)
• 🎨 Multiple styles — Conventional, detailed, emoji, or short
• ✏️ Interactive flow — Preview, edit, regenerate, or confirm before committing
• 🌍 Multi-language — Generate commit messages in any language
• ⚡ Works offline — Ollama support for fully local inference
• 🖥️ Cross-platform — Windows, macOS, and Linux

📦 Installation

# npm (recommended)
npm install -g @ahmad_technology/gitcommit-ai

# Or run directly
npx @ahmad_technology/gitcommit-ai

🚀 Quick Start

# 1. Setup (first time only)
gitcommit setup

# 2. Stage your changes
git add .

# 3. Generate & commit
gitcommit

What you'll see:

⠋ Analyzing 3 changed files...

✨ Suggested commit:

  feat(auth): add OAuth2 login with Google provider

  [ Confirm ]  [ Edit ]  [ Regenerate ]  [ Cancel ]

> Confirm

✅ Committed: feat(auth): add OAuth2 login with Google provider

🔧 Usage

gitcommit                         # Generate from staged diff
gitcommit --all                   # Include unstaged changes
gitcommit --provider anthropic    # Override provider
gitcommit --model claude-opus-4   # Override model
gitcommit --style emoji           # Emoji prefix style
gitcommit --lang fr               # Output in French
gitcommit --dry-run               # Print only, don't commit
gitcommit --copy                  # Copy to clipboard
gitcommit --regenerate            # Get 3 options to choose from
gitcommit setup                   # Interactive config wizard
gitcommit config set provider groq
gitcommit config get defaults.model
gitcommit config list

⚙️ Configuration

Config is stored in ~/.gitcommit/config.toml:

[defaults]
provider = "openai"
model = "gpt-4o"
style = "conventional"
language = "en"
max_diff_lines = 300

[openai]
api_key = "sk-..."

[anthropic]
api_key = "sk-ant-..."

[gemini]
api_key = "AIza..."

[groq]
api_key = "gsk_..."

[openrouter]
api_key = "sk-or-..."

[nim]
api_key = "nvapi-..."
base_url = "https://integrate.api.nvidia.com/v1"

[ollama]
base_url = "http://localhost:11434"
model = "llama3"

🤖 Supported Providers

| Provider | Config Key | Notes | |---|---|---| | OpenAI | openai.api_key | GPT-4o, GPT-4-turbo | | Anthropic | anthropic.api_key | Claude 3.5 Sonnet+ | | Google Gemini | gemini.api_key | gemini-2.0-flash | | Groq | groq.api_key | Llama 3.3, ultra-fast | | NVIDIA NIM | nim.api_key | OpenAI-compatible | | OpenRouter | openrouter.api_key | 100+ models | | Ollama | None | Fully offline |

🛠️ Development

# Clone & install
git clone https://github.com/AhmadTchnology/Ai_Commit_Message_Generator.git
cd Ai_Commit_Message_Generator
npm install

# Run in dev mode
npm run dev

# Run tests
npm test

# Build
npm run build

📄 License

MIT © AhmadTchnology

Made With ❤️ By AhmadTchnology