MCP Server Streamable
A demonstration MCP server that runs over HTTP with streamable transport, providing a simple greeting tool for testing MCP client-server communication. Built with FastMCP framework for educational purposes in learning MCP server development.
README
MCP Server Streamable
A Model Context Protocol (MCP) server implementation using FastMCP with streamable HTTP transport.
Overview
This project demonstrates how to create an MCP server that runs over HTTP with streamable transport. The server provides a simple greeting tool that can be called by MCP clients.
Features
- HTTP Transport: Uses streamable HTTP transport for communication
- FastMCP Framework: Built with the FastMCP library for easy server development
- Simple Tool: Includes a greeting tool that takes a name parameter
Prerequisites
- Python 3.8+
- uv (recommended for fast package management)
Installation
- Clone or navigate to the project directory:
cd d:\work\projects\mcp-servers-course\mcp-server-streamable
- Install dependencies using uv:
uv sync
This will automatically create a virtual environment and install all dependencies from the pyproject.toml file.
- Activate the virtual environment:
uv run python server.py
Or manually activate:
.venv\Scripts\activate
Usage
Running the Server
Start the MCP server:
python server.py
Or using uv:
uv run python server.py
The server will start on the default port and be accessible via HTTP.
Available Tools
greeting
- Description: Send a personalized greeting
- Parameter:
name(string) - The name to include in the greeting - Returns: A greeting message
Example usage:
# When called with name="Alice"
# Returns: "Hi Alice"
Configuration for Claude Desktop
To use this server with Claude Desktop, add the following configuration to your claude_desktop_config.json file:
{
"mcpServers": {
"streamable-server": {
"command": "python",
"args": ["d:\\work\\projects\\mcp-servers-course\\mcp-server-streamable\\server.py"]
}
}
}
Alternatively, if you want to use the remote HTTP endpoint:
{
"mcpServers": {
"streamable-server": {
"command": "npx",
"args": [
"mcp-remote",
"http://127.0.0.1:8000/mcp",
"--allow-http"
]
}
}
}
Development
Project Structure
mcp-server-streamable/
├── server.py # Main server implementation
├── README.md # This file
└── .venv/ # Virtual environment
Extending the Server
To add new tools, define functions with the @mcp.tool() decorator:
@mcp.tool()
def your_tool_name(param: str) -> str:
"""Tool description"""
# Your implementation here
return result
Transport Details
This server uses the "streamable-http" transport, which:
- Provides HTTP-based communication
- Supports streaming responses
- Is suitable for web-based integrations
- Allows for remote access over HTTP
Troubleshooting
- Port conflicts: If the default port is in use, the server will indicate this in the logs
- Virtual environment: Ensure the virtual environment is activated before running
- Dependencies: Make sure all required packages are installed with
uv sync
License
This project is part of the MCP servers course and is intended for educational purposes.
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。
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。
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。
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。