MCP Server Creator
Enables dynamic creation and code generation of MCP servers using FastMCP, with tools for adding custom tools, resources, and generating runnable Python code.
README
MCP Server Creator 🏗️
A powerful Model Context Protocol (MCP) server that creates other MCP servers! This meta-server provides tools for dynamically generating FastMCP server configurations and Python code.
🚀 Quick Start
Install with pip
pip install mcp-server-creator
Run with uvx (recommended)
uvx mcp-server-creator
Run as a module
python -m mcp_server_creator
Features
- Dynamic Server Creation: Create new MCP server configurations on the fly
- Tool Builder: Add custom tools with parameters, return types, and implementations
- Resource Manager: Add static and dynamic resources with template support
- Code Generation: Generate complete, runnable Python code for your servers
- File Export: Save generated servers directly to Python files
- Example Templates: Built-in example server to demonstrate capabilities
Usage
As an MCP Server
The MCP Server Creator is itself an MCP server that can be used with Claude Desktop or any MCP client.
Claude Desktop Configuration
Add to your claude_desktop_config.json:
{
"mcpServers": {
"mcp-server-creator": {
"command": "uvx",
"args": ["mcp-server-creator"]
}
}
}
Programmatic Usage
from mcp_server_creator import create_server, add_tool, generate_server_code
# Create a new server configuration
result = create_server(
name="My API Server",
description="A custom API integration server",
version="1.0.0"
)
# Add a tool
add_tool(
server_id="my_api_server",
tool_name="fetch_data",
description="Fetch data from the API",
parameters=[{"name": "endpoint", "type": "str"}],
return_type="dict",
implementation='return {"data": f"Fetched from {endpoint}"}'
)
# Generate the code
code = generate_server_code("my_api_server")
print(code)
Available Tools
Server Management
create_server: Create a new MCP server configurationlist_servers: List all server configurations in memoryget_server_details: Get detailed information about a specific server
Tool Management
add_tool: Add a tool to an existing server- Supports both sync and async tools
- Custom parameter definitions with types and defaults
- Automatic import management
Resource Management
add_resource: Add a resource to an existing server- Static resources for fixed data
- Dynamic resource templates with parameters
- Custom MIME types
Code Generation
generate_server_code: Generate complete Python code for a serversave_server: Save generated server code to a filecreate_example_server: Create a complete example Weather Service
Example: Creating a Weather Service
import asyncio
from mcp_server_creator import create_server, add_tool, add_resource, save_server
async def create_weather_service():
# Create the server
create_server(
name="Weather Service",
description="Get weather information",
version="1.0.0"
)
# Add a weather tool
add_tool(
server_id="weather_service",
tool_name="get_weather",
description="Get current weather for a city",
parameters=[
{"name": "city", "type": "str"},
{"name": "units", "type": "str", "default": '"celsius"'}
],
return_type="dict",
is_async=True,
implementation='''
# Your weather API logic here
return {
"city": city,
"temperature": 20,
"units": units,
"condition": "sunny"
}'''
)
# Add a resource
add_resource(
server_id="weather_service",
uri="weather://{city}/current",
name="Current Weather",
description="Get current weather data",
is_template=True,
implementation='return {"city": city, "weather": "sunny"}'
)
# Save to file
await save_server("weather_service", "weather_service.py")
# Run the creation
asyncio.run(create_weather_service())
Development
Setup
git clone https://github.com/GongRzhe/mcp-server-creator.git
cd mcp-server-creator
pip install -e .
Testing
python test_mcp_creator.py
Requirements
- Python 3.8+
- FastMCP >= 0.1.0
License
MIT License - see LICENSE file for details.
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
Author
GongRzhe - gongrzhe@gmail.com
Links
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。