🤖 AI Commit Generator (Multi-Provider)

Stop wasting time writing commit messages. Let AI analyze your code changes and generate clear, detailed, and professional commit messages in seconds.

🤔 Why This Extension?

The Problem: Writing good commit messages is tedious. Developers often resort to vague messages like "fix bug" or "update code" because crafting a proper description takes time and mental effort—especially after a long coding session.

The Solution: This extension uses AI to read your actual code changes (git diff) and generate a meaningful, descriptive commit message that follows best practices. No more guessing, no more "WIP" commits.

🎯 Who Is This For?

• 👨‍💻 Solo developers who want cleaner git history without the effort
• 👥 Teams that enforce commit message standards (like Conventional Commits)
• 🎓 Beginners learning how to write proper commit messages
• ⚡ Anyone who values productivity over repetitive tasks

✨ What Does It Do?

🚀 Quick Start

1️⃣ Install

Search "AI Commit Generator" in VSCode Extensions and click Install.

2️⃣ Get a FREE API Key

You need at least one API key (all are FREE):

💡 Pro Tip: Add multiple API keys for better reliability!

3️⃣ Configure

1. Open VSCode Settings (Ctrl + , / Cmd + ,)
2. Search "AI Commit"
3. Paste your API key(s) in the appropriate field(s)

4️⃣ Generate!

1. Make some code changes
2. Click the ✨ sparkle icon in Source Control panel
   • Or press Ctrl+Shift+G then Ctrl+Shift+M
   • Or open Command Palette (Ctrl+Shift+P) → "AI: Generate Commit Message"
3. If files are staged → commit message is generated instantly
4. If no staged files → you'll be prompted:
   • Stage All & Generate — stages everything and generates
   • Use Unstaged Changes — generates without staging
   • Cancel — do nothing
5. Done! Your commit message is ready 🎉

⚙️ Configuration

General Settings

Provider-Specific Settings

Each provider has dedicated settings for:

• 🔑 API Key
• 🤖 Model Selection
• 🌐 API Base URL (optional)

Available Models (as of v1.3.0)

Groq:

• llama-3.3-70b-versatile ⚡ (Recommended - Balanced)
• llama-3.3-70b-specdec 🚀 (Ultra Fast - Speculative Decoding)

OpenRouter:

• arcee-ai/trinity-large-preview:free 🎯 (Best Free)
• arcee-ai/trinity-mini-preview:free ⚡ (Fastest - 2.8s latency)

Mistral:

• codestral-latest 💻 (Best for Code)
• ministral-8b-latest 🪶 (Lightweight - 91.5 tok/s)
• open-mistral-nemo 🔓 (Open Source - 78.6 tok/s)

Cohere:

• command-r-plus 🧠 (Most Capable)
• command-r ⚖️ (Balanced)
• command-r7b-12-2024 🚀 (Fastest - 88.8 tok/s)

📝 Commit Style Examples

🎯 Conventional (Recommended)

feat(auth): implement OAuth2 login flow

- Add Google OAuth provider integration
- Create session management middleware
- Update user model to support external auth

📄 Detailed

Implement OAuth2 login flow with Google provider

This commit adds complete OAuth2 authentication support including
session management and user model updates for external providers.

Changes include:
- Google OAuth2 integration
- Session middleware
- User model extensions

⚡ Simple

Add OAuth2 login with Google provider

🔒 Privacy & Security

• ✅ Your API keys are stored locally on your machine (in VSCode settings)
• ✅ Code diffs are sent directly to the AI provider you choose
• ✅ The extension developer has NO access to your keys or code
• ✅ Open source — inspect the code yourself on GitHub
• ✅ No telemetry or tracking — your privacy is respected

🐛 Troubleshooting

❌ "No API key configured"

Solution: Add at least one API key in VSCode Settings → search "AI Commit"

❌ "Failed to generate commit message"

Solutions:

1. Check if you have staged changes (git add .)
2. Verify your API key is valid
3. Try a different provider (set Preferred Provider to another option)
4. Check internet connection

❌ Rate limit errors

Solution: The extension will automatically try the next provider. Consider adding multiple API keys for better reliability.

⚠️ Slow generation

Solutions:

1. Use faster models (e.g., Groq's llama-3.3-70b-specdec)
2. Switch to lightweight models (e.g., Mistral's ministral-8b-latest)
3. Check your internet speed

🛠️ For Developers

Want to contribute or run locally?

# Clone repository
git clone https://github.com/muhammadalifh/vscode-ai-commit.git
cd vscode-ai-commit

# Install dependencies
npm install

# Compile TypeScript
npm run compile

# Launch Extension Development Host
# Press F5 in VSCode

Project Structure

vscode-ai-commit/
├── src/
│   ├── extension.ts          # Main entry point
│   ├── providers/             # AI provider implementations
│   ├── services/              # Core services
│   └── utils/                 # Helper functions
├── package.json              # Extension manifest
└── tsconfig.json             # TypeScript config

Contributing

See CONTRIBUTING.md for guidelines.

📊 Performance Benchmarks

Benchmarks based on average response time for 200-line diffs

🔄 Changelog

See CHANGELOG.md for version history.

📝 License

MIT License — use it freely, even commercially.

See LICENSE for full text.

🙏 Credits & Acknowledgments

Built with ❤️ by Muhammad Alif H

Powered By:

• Groq — Lightning-fast LLaMA 3.3 70B inference
• OpenRouter — Unified API for multiple models
• Mistral AI — Open-source excellence
• Cohere — Enterprise-grade language models

Special Thanks:

• VSCode Extension API team
• The open-source community
• All contributors and users

💬 Support & Feedback

• 🐛 Bug Reports: GitHub Issues
• 💡 Feature Requests: GitHub Discussions
• ⭐ Like it? Star the repo and leave a review!

🌟 Star History

Made with 💻 and ☕ by developers, for developers.