lazy-commit 0.1.8

A git commit message generator for personal use.

lazy-commit

A smart git commit message generator that uses AI to create high-quality commit messages following the Conventional Commits specification. It analyzes your git changes and automatically generates descriptive commit messages.

Features

• 🤖 AI-Powered: Uses OpenAI-compatible APIs to generate intelligent commit messages
• 📝 Conventional Commits: Follows the Conventional Commits specification format
• 🎯 Smart File Analysis: Automatically analyzes changed files and generates appropriate diffs
• 🚫 Intelligent Filtering: Excludes lock files and binary files from analysis while still including them in commits
• ⚡ Context Compression: Automatically compresses large diffs to fit within token limits
• 🔄 Auto-Push Support: Optional automatic push to remote repository after commit
• 🎨 Rich UI: Beautiful command-line interface with progress bars and status indicators
• 📊 File Status Display: Shows modified, added, and deleted files with clear indicators

Installation

Using uv (recommended)

uv tool install lazy-commit

uv tool upgrade lazy-commit

From source

git clone https://github.com/QIN2DIM/lazy-commit.git
cd lazy-commit
uv sync

Configuration

Set the required environment variables:

# For OpenAI API
export LAZY_COMMIT_OPENAI_BASE_URL="https://api.openai.com/v1"
export LAZY_COMMIT_OPENAI_API_KEY="your-openai-api-key"
export LAZY_COMMIT_OPENAI_MODEL_NAME="gpt-4o-mini"

# For free models via OpenRouter
export LAZY_COMMIT_OPENAI_BASE_URL="https://openrouter.ai/api/v1"
export LAZY_COMMIT_OPENAI_API_KEY="sk-or-v1-xxx"
export LAZY_COMMIT_OPENAI_MODEL_NAME="moonshotai/kimi-k2:free"

# For Chinese users - free models via ModelScope
export LAZY_COMMIT_OPENAI_BASE_URL="https://api-inference.modelscope.cn/v1"
export LAZY_COMMIT_OPENAI_API_KEY="ms-xxx"
export LAZY_COMMIT_OPENAI_MODEL_NAME="Qwen/Qwen3-30B-A3B-Instruct-2507"

# Optional: Set maximum context size (default: 32000)
export LAZY_COMMIT_MAX_CONTEXT_SIZE=32000

# Optional: Bypass system proxy for LAN endpoints (default: false)
# Useful when accessing internal model endpoints via VPN
export LAZY_COMMIT_BYPASS_PROXY=true

Environment File

You can also create a .env file in your project root:

For OpenRouter (free models):

LAZY_COMMIT_OPENAI_BASE_URL=https://openrouter.ai/api/v1
LAZY_COMMIT_OPENAI_API_KEY=sk-or-v1-xxx
LAZY_COMMIT_OPENAI_MODEL_NAME=moonshotai/kimi-k2:free

For Chinese users - ModelScope (free models):

LAZY_COMMIT_OPENAI_BASE_URL="https://api-inference.modelscope.cn/v1"
LAZY_COMMIT_OPENAI_API_KEY="ms-xxx"
LAZY_COMMIT_OPENAI_MODEL_NAME="Qwen/Qwen3-30B-A3B-Instruct-2507"

Usage

Basic Usage

Generate a commit message only (display message without applying):

commit

or if installed from source:

uv run commit

Stage and Commit

Generate commit message, stage files, and apply commit (without push):

commit --add

Auto-push After Commit

Generate commit message, stage files, apply commit, and push to remote:

commit --push

Note: When --push is enabled, --add is automatically enabled.

How It Works

1. Repository Detection: Automatically detects git repository root
2. File Analysis: Scans for modified, staged, and untracked files
3. Smart Filtering: Excludes files like *.lock, *.ipynb from AI analysis but includes them in commits
4. Diff Generation: Creates comprehensive diffs for all relevant changes
5. Context Management: Compresses large diffs to fit within AI model context limits
6. AI Generation: Uses AI to generate a structured commit message following Conventional Commits
7. Display Message: Always displays the generated commit message
8. Optional Staging & Commit: If --add is used, stages all changes and applies the generated commit message
9. Optional Push: If --push is used, pushes changes to remote repository after commit

Commit Message Format

The tool generates commit messages following the Conventional Commits specification:

<type>(<scope>): <description>

[optional body]

[optional footer]

Commit Types

• feat: New features
• fix: Bug fixes
• docs: Documentation changes
• style: Code style changes (formatting, etc.)
• refactor: Code refactoring
• perf: Performance improvements
• test: Test additions or modifications
• build: Build system changes
• ci: CI/CD changes
• chore: Maintenance tasks

Examples

Feature Addition

feat(api): add user authentication endpoint

Implements JWT-based authentication with login and logout functionality.
Includes input validation and error handling for invalid credentials.

Bug Fix

fix(payment): correct tax calculation error

Fixed off-by-one error in tax calculation loop that was causing
incorrect tax amounts for orders with multiple items.

Documentation

docs(README): update installation and configuration instructions

Added detailed setup guide for environment variables and
included examples for different AI providers.

Requirements

• Python 3.12+
• Git repository
• OpenAI-compatible API access (OpenAI, local models, etc.)

Dependencies

• typer: Command-line interface
• openai: OpenAI API client
• pydantic-settings: Configuration management
• rich: Beautiful terminal output
• loguru: Logging
• tiktoken: Token counting

Development

Setup Development Environment

git clone https://github.com/QIN2DIM/lazy-commit.git
cd lazy-commit
uv sync --group dev

Code Quality

# Format code
uv run black src/

# Lint code
uv run ruff check src/

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Run code quality checks
6. Submit a pull request

License

This project is licensed under the MIT License. See the LICENSE file for details.