https://github.com/fayazk/unichat-ai
A unified desktop interface for multiple AI models with MCP integration, organized project management, and secure API handling
https://github.com/fayazk/unichat-ai
ai-chat chatgpt claude desktop-app electron llm mcp mcp-client ollama openai
Last synced: 4 months ago
JSON representation
A unified desktop interface for multiple AI models with MCP integration, organized project management, and secure API handling
- Host: GitHub
- URL: https://github.com/fayazk/unichat-ai
- Owner: FayazK
- Created: 2025-03-13T05:00:52.000Z (7 months ago)
- Default Branch: main
- Last Pushed: 2025-04-16T04:02:46.000Z (6 months ago)
- Last Synced: 2025-04-16T04:41:58.605Z (6 months ago)
- Topics: ai-chat, chatgpt, claude, desktop-app, electron, llm, mcp, mcp-client, ollama, openai
- Language: JavaScript
- Homepage:
- Size: 14.4 MB
- Stars: 2
- Watchers: 1
- Forks: 1
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# UniChat AI
UniChat AI is a cross-platform desktop application that provides a unified interface for interacting with multiple large language models (LLMs) through a single elegant chat interface. With built-in support for Model Context Protocol (MCP), UniChat AI extends the capabilities of AI models with tools, resources, and custom integrations.
## Features
- 🤖 **Multi-Model Support**: Chat with models from OpenAI, Anthropic, Google Gemini, and local models via Ollama
- 📂 **Project Management**: Organize conversations into projects with custom settings and instructions
- 🔌 **MCP Integration**: Connect to any Model Context Protocol server to extend AI capabilities
- 🔄 **Continuous Context**: Switch models mid-conversation while maintaining context
- 📎 **File Attachments**: Attach and reference files in your conversations
- 🔐 **Secure API Management**: Securely store your API keys in your system's credential store
- 🌓 **Light/Dark Modes**: Work comfortably day or night with theme support
- 💾 **Data Ownership**: All your conversations are stored locally
## Installation
### Download pre-built binaries
Download the latest release for your platform from the [Releases](https://github.com/fayazk/unichat-ai/releases) page.
| Platform | Download |
|----------|----------|
| Windows | [UniChat-AI-Windows.exe](https://github.com/fayazk/unichat-ai/releases) |
| macOS | [UniChat-AI-macOS.dmg](https://github.com/fayazk/unichat-ai/releases) |
| Linux | [UniChat-AI-Linux.AppImage](https://github.com/fayazk/unichat-ai/releases) |
### Build from source
```bash
# Clone the repository
git clone https://github.com/fayazk/unichat-ai.git
cd unichat-ai
# Install dependencies
npm install
# Run in development mode
npm run dev
# Build for production
npm run build
```
## Getting Started
1. **Launch the application** after installation
2. **Add your API keys** in the Settings panel
3. **Create your first project** by clicking the "+" button in the sidebar
4. **Start chatting** with your preferred AI model
5. **Configure MCP servers** (optional) to extend AI capabilities
## MCP Integration
UniChat AI supports Model Context Protocol (MCP) servers for extending AI capabilities with tools and resources. To configure MCP servers:
1. Navigate to Settings > MCP Servers
2. Add a new server configuration:
```json
{
"name": "filesystem",
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/directory"]
}
```
Learn more about available MCP servers in the [MCP documentation](https://modelcontextprotocol.io/examples).
## API Keys
UniChat AI requires API keys to access different LLM providers:
- **OpenAI**: [Get API key](https://platform.openai.com/api-keys)
- **Anthropic**: [Get API key](https://console.anthropic.com/settings/keys)
- **Google Gemini**: [Get API key](https://ai.google.dev/tutorials/setup)
API keys are securely stored in your system's credential store and never shared.
## For Developers
### Architecture
UniChat AI uses a modular architecture with these key components:
1. **UI Layer**: Electron with React/Vue components
2. **Provider Layer**: Adapters for different LLM APIs
3. **MCP Layer**: Integration with MCP servers
4. **Storage Layer**: Local database and file system interaction
### Adding New LLM Providers
To add support for a new LLM provider:
1. Create a new adapter in `src/api/providers/`
2. Implement the `ProviderInterface`
3. Register the provider in `src/api/providerRegistry.ts`
See the [Developer Guide](docs/DEVELOPMENT.md) for detailed instructions.
## Troubleshooting
### Common Issues
- **API Connection Issues**: Verify your API keys and internet connection
- **MCP Server Not Connecting**: Check your MCP server configuration and ensure the correct paths
- **Missing Messages**: If conversations disappear, check the application logs for database errors
For more help, see the [Troubleshooting Guide](docs/TROUBLESHOOTING.md) or open an issue.
## Contributing
Contributions are welcome! Please read our [Contributing Guidelines](CONTRIBUTING.md) before submitting a pull request.
## License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## Acknowledgements
- [Electron](https://www.electronjs.org/)
- [Model Context Protocol](https://modelcontextprotocol.io/)
- [OpenAI](https://openai.com/)
- [Anthropic](https://www.anthropic.com/)
- [Google Gemini](https://deepmind.google/technologies/gemini/)
- [Ollama](https://ollama.ai/)
---
Made with ❤️ by Fayaz K
[GitHub](https://github.com/fayazk) | [Website](https://fayazk.com)