Simple MCP Client

@guolisen

Basic information

About Simple MCP Client

A simple command-line MCP (Model Context Protocol) client for testing MCP servers using Python. It supports chatting with various LLM providers and connecting to multiple MCP servers.

Simple MCP Client config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "simple_mcp_client": {
      "command": "uv",
      "args": [
        "venv"
      ]
    }
  }
}

Simple MCP Client tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Simple MCP Client?

Simple MCP Client is a Python command-line client for testing MCP (Model Context Protocol) servers. It supports chatting with various LLM providers and connecting to multiple MCP servers via SSE and stdio protocols, making it ideal for developers building and debugging MCP server integrations.

How to use Simple MCP Client?

Install via pip install simple-mcp-client or use UV, then configure a JSON file at ~/.config/simple_mcp_client/config.json (Linux/macOS) or %APPDATA%\simple_mcp_client\config.json (Windows). Start the client by running simple-mcp-client or python -m simple_mcp_client.main. Use commands like connect, chat, execute, and help to interact with MCP servers and LLM models.

Key features of Simple MCP Client

Connect to multiple MCP servers (SSE and stdio)
Integrate with Ollama, DeepSeek, OpenAI, and OpenRouter
Interactive command-line interface with auto-completion
Chat mode with ReAct agent for intelligent tool use
Execute tools directly with execute command
View available tools, resources, and prompts

Use cases of Simple MCP Client

Test and debug individual MCP servers interactively
Chain multiple MCP server tools for multi-step problem solving
Build prototypes that combine LLM reasoning with server tool execution
Explore available resources and prompts from connected MCP servers
Automate tool execution in development workflows

Frequently asked questions

Where is the configuration file stored?

On Linux/macOS it is stored at `~/.config/simple_mcp_client/config.json`; on Windows it is at `%APPDATA%\simple_mcp_client\config.json`. The client copies a default configuration if the file does not exist.

What are the runtime requirements?

Python 3.12 or higher is required. The client uses additional dependencies listed in the project's `requirements.txt`.

How do I connect to an MCP server?

Use the `connect <server_name>` command. The server must be defined in the configuration file under `mcpServers`, specifying its type (`sse` or `stdio`) and connection details.

What LLM providers are supported?

The client supports Ollama, DeepSeek, OpenAI, and OpenRouter. The provider and model are set in the configuration file under `llm`.

How does authentication work for LLM providers?

Authentication is handled by setting an `api_key` in the `llm` section of the configuration file. Some providers may also require an `api_url`.

Comments

More Other MCP servers

ICSS

chokcoco

不止于 CSS

20.7K

Yuppsy

Yuppsy MCP: build and publish real estate websites from your AI assistant. Yuppsy lets your AI assistant create live websites for real estate agents and individual property listings. You don't need a designer or any code

FastMCP v2 🚀

jlowin

🚀 The fast, Pythonic way to build MCP servers and clients.

26.0K

Senso MCP for Joomla

sensomedia

Senso MCP turns a Joomla 5 or 6 site into a remote MCP server. It runs inside Joomla itself (no extra server, no Node.js bridge) and is secured with OAuth 2.1 and Joomla's own ACL. The free pack lets an AI assistant crea

EverArt

modelcontextprotocol

Model Context Protocol Servers

88.1K