CLI MCP Chat

A command-line MCP (Model Context Protocol) client that uses OpenAI instead of Anthropic Claude, allowing you to connect to multiple MCP servers and interact with them through OpenAI's chat models.

Features

Multi-server support: Connect to multiple MCP servers simultaneously
OpenAI integration: Uses OpenAI models instead of Anthropic Claude
Streaming responses: Real-time streaming from OpenAI API
Flexible configuration: Supports custom OpenAI base URLs for self-hosted or compatible endpoints
Environment-based setup: Configure all settings via environment variables

Installation

Clone this repository:

git clone https://github.com/SkunkApe-jp/cli-mcp-chat.git
cd cli-mcp-chat

Install dependencies:

pip install -r requirements.txt

Configuration

Copy the example environment file:

cp .env.example .env

Edit .env and add your OpenAI API credentials:

# REQUIRED: your OpenAI API key
OPENAI_API_KEY=sk-your-api-key

# REQUIRED: the OpenAI model name to use (chat + tools capable)
# Examples: gpt-4.1, gpt-4.1-mini, gpt-4.1-preview, etc.
OPENAI_MODEL=gpt-4.1-mini

# OPTIONAL: override the base URL (for self-hosted / compatible endpoints)
# For the official OpenAI API you can leave this unset.
# Example: https://api.openai.com/v1
OPENAI_BASE_URL=

Usage

Run the client:

python client.py

The client will start and allow you to interact with connected MCP servers using OpenAI's models.

Requirements

Python 3.8+
OpenAI API key
MCP servers to connect to

Dependencies

openai>=1.0.0 - OpenAI API client
python-dotenv>=1.0.0 - Environment variable management
mcp>=1.0.0 - Model Context Protocol client

License

This project is open source and available under the MIT License.

MCP Servers