simple-mcp-server
A simple MCP server offering three utility tools: UUID generation, temperature conversion, and text statistics.
README
Simple MCP Server
A simple Model Context Protocol (MCP) server built with FastMCP, featuring three utility tools.
Tools
- generate_uuid - Generates a random UUID (version 4 or 1)
- convert_temperature - Converts temperature between Celsius and Fahrenheit
- text_statistics - Calculates basic text statistics (characters, words, lines)
Installation
Install dependencies using uv:
uv sync
Usage
Running the Server
Start the MCP server with HTTP transport:
uv run server.py
The server will start on http://localhost:8000/mcp by default.
Testing with FastMCP Client
Run the FastMCP-based client:
uv run client_fastmcp.py
Testing with httpx Client
Run the pure httpx client (no MCP libraries required):
uv run client_httpx.py
Testing with curl Client
Run the bash script using only curl and jq:
./client_curl.sh
Requirements for curl client:
curl(for HTTP requests)jq(for JSON parsing)- Standard Unix tools (
grep,sed,awk,tr)
Project Structure
server.py- MCP server with three utility toolsclient_fastmcp.py- Client using FastMCP libraryclient_httpx.py- Client using only httpx for direct HTTP requestsclient_curl.sh- Bash script client using curl and jqmain.py- Original template file (not used)
Example Output
All three clients will:
- Connect to the server
- List available tools
- Test generate_uuid (generates a UUID v4)
- Test convert_temperature (converts 0°C to Fahrenheit → 32.0°F)
- Test text_statistics (counts characters, words, and lines in sample text)
Example:
Test 1: Generate UUID
Result: 98f99e94-7357-4dd4-85eb-76819fad3ada
Test 2: Convert temperature (0°C to Fahrenheit)
Result: 32.0
Test 3: Text statistics
Result: {"characters":47,"words":9,"lines":3}
Technical Details
- Transport: Streamable HTTP (MCP protocol over HTTP)
- Endpoint:
/mcp - Response Format: Server-Sent Events (SSE)
- Session Management: Server assigns session ID via
Mcp-Session-Idheader
Client Implementations
-
FastMCP Client (
client_fastmcp.py):- Uses official MCP Python SDK
- Handles SSE automatically
- Full protocol support
-
httpx Client (
client_httpx.py):- Pure Python with only httpx dependency
- Manual JSON-RPC implementation
- Custom SSE parsing
-
curl Client (
client_curl.sh):- Bash script using curl and jq
- Demonstrates raw HTTP requests
- No Python dependencies required
Requirements
- Python 3.13+
- fastmcp
- httpx (included as fastmcp dependency)
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。