drio
Open app

mcpx-py

Source

Python client library for

Catalog onlyCatalog onlySTDIO

Overview

mcpx-py is a Python client library designed for interacting with large language models (LLMs) using mcp.run tools, enabling seamless integration with various AI providers.

To use mcpx-py, install it via pip or uv, set up your mcp.run session, and then utilize the library to interact with different AI models through a chat interface or command line.

  • Supports multiple AI providers including OpenAI, Claude, and Gemini. - Real-time chat interface for interactive conversations with AI models. - Tool suggestion and execution within conversations. - Local and cloud-based AI provider support.
  1. Building chatbots that leverage advanced AI models.
  2. Automating tasks through AI-driven commands.
  3. Integrating AI capabilities into Python applications.

Add to your AI client

Use these steps to connect mcpx-py in Cursor, Claude, VS Code, and other MCP-compatible apps. The same JSON appears in the Use with menu above for one-click copy.

Cursor

Add this to your .cursor/mcp.json file in your project root, then restart Cursor.

.cursor/mcp.json

{
  "mcpServers": {
    "mcpx-py-dylibso": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcpx-py-dylibso"
      ]
    }
  }
}

Claude Desktop

Add this server entry to the mcpServers object in your Claude Desktop config, then restart the app.

~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows)

{
  "mcpServers": {
    "mcpx-py-dylibso": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcpx-py-dylibso"
      ]
    }
  }
}

Claude Code

Add this to your project's .mcp.json file. Claude Code will detect it automatically.

.mcp.json (project root)

{
  "mcpServers": {
    "mcpx-py-dylibso": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcpx-py-dylibso"
      ]
    }
  }
}

VS Code (Copilot)

Add this to your .vscode/mcp.json file. Requires the GitHub Copilot extension with MCP support enabled.

.vscode/mcp.json

{
  "servers": {
    "mcpx-py-dylibso": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcpx-py-dylibso"
      ]
    }
  }
}

Windsurf

Add this to your Windsurf MCP config file, then restart Windsurf.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "mcpx-py-dylibso": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcpx-py-dylibso"
      ]
    }
  }
}

Cline

Open Cline settings, navigate to MCP Servers, and add this server configuration.

Cline MCP Settings (via UI)

{
  "mcpServers": {
    "mcpx-py-dylibso": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-mcpx-py-dylibso"
      ]
    }
  }
}

FAQ

**What AI providers are supported?**

mcpx-py supports several AI providers including OpenAI, Claude, Ollama, and Gemini.

**Is mcpx-py free to use?**

Yes! mcpx-py is open-source and free to use.

**How do I set up my environment for mcpx-py?**

You need to install the library, generate a session ID using mcp.run, and set the necessary environment variables for your chosen AI provider.

Developer tools