MCP-BOS: 可扩展的MCP服务器框架

Name: MCP-BOS: 可扩展的MCP服务器框架
Author: gooboot

@gooboot

About MCP-BOS: 可扩展的MCP服务器框架

MCP-BOS: 模块化、可扩展的Model Context Protocol服务器框架使用基于约定的自动模块发现机制，为Claude Desktop打造的灵活MCP服务器框架。通过简洁的模块接口和声明式配置，轻松扩展AI应用功能，无需修改核心代码。支持FastMCP标准，包含完整工具、资源和提示模板注册能力。

#mcp #python3 #claude #mcp-server #mcp-client #fastmcp

Basic information

Config

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

{
  "mcpServers": {
    "MCP-BOS": {
      "command": "uv",
      "args": [
        "pip",
        "install",
        "mcp[cli]"
      ]
    }
  }
}

Tools

No tools detected

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

Overview

What is MCP-BOS?

MCP-BOS is a modular, extensible Model Context Protocol (MCP) server framework that uses convention‑based automatic module discovery. Designed for Claude Desktop, it provides a flexible way to extend AI application capabilities without modifying core code, supporting the FastMCP standard with full tool, resource, and prompt template registration.

How to use MCP-BOS?

Install dependencies with uv pip install mcp[cli], then edit the config.json file to configure global settings and modules. Run directly with python main.py or uv run main.py. Integrate with Claude Desktop using mcp install main.py, or test with mcp inspect main.py.

Key features of MCP-BOS

Modular design: functions organized as self‑contained modules for easy extension
Automatic discovery: convention‑over‑configuration module loading
Declarative configuration: flexible setup via config.json
Plug‑and‑play: new features added by creating a module directory
Stable architecture: layered design keeps core secure and reliable
Detailed logging: comprehensive logs for debugging and monitoring
Claude Desktop integration: deep integration for AI‑enhanced experiences

Use cases of MCP-BOS

Extend Claude Desktop with custom tools, resources, or prompt templates
Build modular AI assistants with independently developed modules
Rapidly prototype MCP servers using the pre‑built framework
Manage multiple functionality modules through declarative configuration

FAQ from MCP-BOS

Why is my module not loading?

Check that the module directory structure is correct, that __init__.py properly exports the module class, that the module is enabled in config.json, and review logs for detailed error information.