Kiters MCP Client
- agent-framework
About
What is Kiters MCP Client?
An Electron-based desktop chat interface that connects to MCP (Model Context Protocol) servers and integrates with OpenRouter.ai for AI chat. It allows users to view and call tools from MCP servers while conversing with AI models.
How to use Kiters MCP Client?
Clone the repository, install dependencies with npm install, create a .env file with your OpenRouter.ai API key, then run npm run dev for development or npm run build && npm run build:electron to produce a distributable package. Inside the app, add MCP servers via local stdio or remote HTTP, connect to them, select an AI model from the dropdown, and start chatting.
Key features of Kiters MCP Client
- Connect to MCP servers via stdio (local) or HTTP (remote)
- View and call available tools from connected MCP servers
- Chat with AI models from OpenRouter.ai
- Tool calling support for MCP server tools
- Dark/light theme support
Use cases of Kiters MCP Client
- Interact with custom MCP servers in a chat interface
- Build AI assistants that can call external tools via MCP
- Test and debug MCP servers through a graphical interface
- Experiment with various OpenRouter.ai models connected to MCP tools
FAQ from Kiters MCP Client
What is a MCP server?
MCP stands for Model Context Protocol. This client connects to MCP servers to view and call their tools, allowing AI models to use those tools during conversation.
How do I connect a MCP server?
Click the "+" button in the Servers section, enter a name, select server type (Local for stdio or Remote for HTTP), provide the path or URL, then click "Add Server" and press "Connect". Once connected, the available tools are displayed.
Can I use my own API key?
Yes, you need an OpenRouter.ai API key set in a .env file or in
Details
- Author
- namank5x
- GitHub stars
- 1
- Category
- agent-framework
- Repository
- namank5x/kiters-mcp-client
Kiters MCP Client
A simple Electron-based chat interface client for Model Context Protocol (MCP) servers with OpenRouter.ai integration.
Features
- Connect to MCP servers via stdio (local) or HTTP (remote)
- View and call available tools from connected MCP servers
- Chat with AI models from OpenRouter.ai
- Tool calling support for MCP server tools
- Dark/light theme support
Prerequisites
- Node.js (v16+)
- npm or yarn
Installation
1. Clone the repository:
git clone https://github.com/yourusername/kiters-mcp-client.git
cd kiters-mcp-client
2. Install dependencies:
npm install
3. Create a .env file in the root directory and add your OpenRouter.ai API key:
OPENROUTER_API_KEY=your_api_key_here
Development
To start the development server:
npm run dev
This will start both the Electron app and the React development server.
Building
To build the application:
npm run build
npm run build:electron
This will create a distributable package in the dist directory.
Using the Application
Adding an MCP Server
1. Click the "+" button in the Servers section
2. Enter a name for the server
3. Select the server type:
- Local (stdio): Provide the path to the server script (e.g., /path/to/server.js or /path/to/server.py)
- Remote (HTTP): Provide the URL to the server (e.g., http://localhost:3001)
4. Click "Add Server"
Connecting to an MCP Server
1. Click the "Connect" button next to the server you want to connect to
2. Once connected, the available tools will be displayed
Chatting with AI
1. Select a model from the dropdown at the top
2. Type your message in the input field
3. Press Enter or click the Send button
4. The AI will respond and may call tools from the connected MCP server if needed
Settings
Click the "Settings" button at the bottom of the sidebar to:
- Set your OpenRouter.ai API key
- Change the theme (Light, Dark, or System)
License
MIT
