发现优秀的 MCP 服务器

通过 MCP 服务器扩展您的代理能力,拥有 18,836 个能力。

全部18,836
PrestaShop MCP Server

PrestaShop MCP Server

Provides instant, offline access to comprehensive PrestaShop development documentation including 647+ hooks, module guides, component architecture, APIs, and theme development resources for AI assistants.

Crowdlistening

Crowdlistening

Crowdlistening

Uncover MCP

Uncover MCP

发现 MCP 服务器 (Fāxiàn MCP fúwùqì)

OmniFocus MCP Server

OmniFocus MCP Server

Claude 的 OmniFocus 集成:让 LLM 通过模型上下文协议与您的任务交互。使用自然语言命令添加、组织和查询您的 OmniFocus 数据库。

writers-muse-mcp

writers-muse-mcp

一个MCP服务器,用于分析作者的写作风格并生成一篇博文。 (Alternatively, more literally: 一个分析作者写作风格并生成博文的MCP服务器。)

my-mcp-server

my-mcp-server

lilo-vacation-rentals

lilo-vacation-rentals

The only MCP server with AI guest risk scoring and extortion detection. Search properties, book instantly, protect hosts. 41 tools across 5 layers. Instant API key, 10 free credits.

🌟 Unsplash MCP Server Repository

🌟 Unsplash MCP Server Repository

🔎 用于 Unsplash 图片搜索的 MCP 服务器。

Superprecio MCP Server

Superprecio MCP Server

Enables AI assistants to search products, compare prices, and find the best deals across multiple supermarkets in Argentina through Superprecio's price comparison API. Transforms Claude into an expert shopping assistant for Latin American grocery shopping.

ws-mcp

ws-mcp

MCP Servers Multi-Agent AI Infrastructure

MCP Servers Multi-Agent AI Infrastructure

EPSS MCP Server

EPSS MCP Server

A server that retrieves CVE details from the NVD API and fetches EPSS scores to provide comprehensive vulnerability information, including descriptions, CWEs, CVSS scores, and exploitation likelihood percentiles.

qBittorrent MCP

qBittorrent MCP

A service that provides programmatic access to qBittorrent's WebUI API, enabling management of torrents, trackers, tags, speed controls, and system information through natural language.

Minimal MCP Server

Minimal MCP Server

好的,以下是一个 Model Context Protocol (MCP) 服务器的最小实现,用 Python 编写,并使用 `asyncio` 库进行异步操作。 这个例子展示了如何监听连接,接收消息,并发送简单的响应。 ```python import asyncio import json async def handle_client(reader, writer): """处理单个客户端连接.""" addr = writer.get_extra_info('peername') print(f"连接来自 {addr}") try: while True: data = await reader.read(1024) # 读取最多 1024 字节 if not data: break message = data.decode() print(f"接收到消息: {message}") try: # 尝试解析 JSON request = json.loads(message) # 在这里添加你的 MCP 逻辑 # 例如,根据请求类型执行不同的操作 if "method" in request: method = request["method"] if method == "ping": response = {"result": "pong"} elif method == "get_model_info": response = {"result": {"model_name": "MyModel", "version": "1.0"}} # 示例模型信息 else: response = {"error": "Unknown method"} else: response = {"error": "Invalid request"} response_json = json.dumps(response) writer.write(response_json.encode()) await writer.drain() # 刷新缓冲区 print(f"发送响应: {response_json}") except json.JSONDecodeError: print("接收到无效的 JSON") writer.write(b'{"error": "Invalid JSON"}') await writer.drain() except ConnectionError as e: print(f"连接错误: {e}") finally: print(f"关闭连接 {addr}") writer.close() await writer.wait_closed() async def main(): """启动服务器.""" server = await asyncio.start_server( handle_client, '127.0.0.1', 8888) # 监听本地地址 127.0.0.1 端口 8888 addr = server.sockets[0].getsockname() print(f'服务于 {addr}') async with server: await server.serve_forever() if __name__ == "__main__": asyncio.run(main()) ``` **代码解释:** 1. **`handle_client(reader, writer)` 函数:** - 这是处理每个客户端连接的主要函数。 - `reader` 和 `writer` 是 `asyncio.StreamReader` 和 `asyncio.StreamWriter` 对象,用于读取和写入数据。 - `writer.get_extra_info('peername')` 获取客户端的地址。 - `reader.read(1024)` 从客户端读取最多 1024 字节的数据。 - `data.decode()` 将接收到的字节数据解码为字符串。 - `json.loads(message)` 尝试将接收到的消息解析为 JSON 对象。 - **MCP 逻辑:** 这是代码的核心部分,你需要根据 MCP 协议的要求实现你的逻辑。 在这个例子中,它检查请求中是否存在 "method" 字段,并根据方法名称返回不同的响应。 `ping` 方法返回 `pong`,`get_model_info` 返回一个示例模型信息。 - `json.dumps(response)` 将响应转换为 JSON 字符串。 - `writer.write(response_json.encode())` 将 JSON 字符串编码为字节数据并发送给客户端。 - `await writer.drain()` 刷新缓冲区,确保数据被发送。 - `writer.close()` 关闭连接。 - `await writer.wait_closed()` 等待连接完全关闭。 - 异常处理:包括 `json.JSONDecodeError` 处理无效的 JSON,以及 `ConnectionError` 处理连接错误。 2. **`main()` 函数:** - `asyncio.start_server(handle_client, '127.0.0.1', 8888)` 启动一个 TCP 服务器,监听本地地址 127.0.0.1 的 8888 端口。 `handle_client` 函数将处理每个新的客户端连接。 - `server.serve_forever()` 运行服务器,直到手动停止。 3. **`if __name__ == "__main__":` 块:** - 确保 `asyncio.run(main())` 只在脚本直接运行时执行,而不是作为模块导入时执行。 **如何运行:** 1. **保存代码:** 将代码保存为 `mcp_server.py`。 2. **运行脚本:** 在终端中运行 `python mcp_server.py`。 **如何测试:** 你可以使用 `telnet` 或 `netcat` 等工具来连接服务器并发送 MCP 请求。 例如: ```bash telnet 127.0.0.1 8888 ``` 然后,你可以发送一个 JSON 请求,例如: ```json {"method": "ping"} ``` 服务器应该返回: ```json {"result": "pong"} ``` 或者发送: ```json {"method": "get_model_info"} ``` 服务器应该返回: ```json {"result": {"model_name": "MyModel", "version": "1.0"}} ``` **重要注意事项:** * **错误处理:** 这个例子只包含基本的错误处理。 在实际应用中,你需要添加更完善的错误处理机制,例如记录错误日志,返回更详细的错误信息给客户端。 * **安全性:** 这个例子没有考虑安全性。 在生产环境中,你需要采取安全措施,例如身份验证、授权、加密等。 * **MCP 协议:** 这个例子只是一个框架。 你需要根据 MCP 协议的规范来实现你的 MCP 逻辑。 这包括定义消息格式、方法名称、参数、返回值等。 * **异步编程:** `asyncio` 是一个强大的异步编程库。 如果你不熟悉 `asyncio`,建议先学习一下 `async` 和 `await` 的用法。 * **依赖:** 这个例子只依赖于 Python 的标准库 `asyncio` 和 `json`,不需要安装额外的依赖。 * **扩展性:** 这个例子是一个最小实现。 你可以根据需要扩展它,例如添加多线程支持、使用更高效的序列化方法、支持更多的 MCP 方法等。 **下一步:** 1. **定义你的 MCP 协议:** 明确你的 MCP 协议的规范,包括消息格式、方法名称、参数、返回值等。 2. **实现你的 MCP 逻辑:** 根据你的 MCP 协议,实现 `handle_client` 函数中的 MCP 逻辑。 3. **添加错误处理:** 添加更完善的错误处理机制。 4. **考虑安全性:** 采取安全措施,例如身份验证、授权、加密等。 5. **测试和调试:** 充分测试和调试你的服务器。 这个最小实现为你提供了一个起点。 你可以根据你的具体需求来扩展和修改它。

MCP Weather Server

MCP Weather Server

Enables AI agents to access real-time and historical weather data through multiple weather APIs including OpenMeteo, Tomorrow.io, and OpenWeatherMap. Provides comprehensive meteorological information including current conditions, forecasts, historical data, and weather alerts.

46elks MCP Server

46elks MCP Server

Enables AI assistants and MCP-compatible clients to send and manage SMS messages through the 46elks API, leveraging Swedish telecommunications infrastructure.

FastAPI MCP-Style Server

FastAPI MCP-Style Server

A minimal FastAPI implementation that mimics Model Context Protocol functionality with JSON-RPC 2.0 support. Provides basic tools like echo and text transformation through both REST and RPC endpoints for testing MCP-style interactions.

Claude Code DingTalk MCP Server

Claude Code DingTalk MCP Server

Integrates Claude Code with DingTalk (钉钉) robot notifications, allowing users to send task completion alerts and various message formats to DingTalk groups from Claude Code.

Bear MCP Server

Bear MCP Server

Enables AI assistants to search, read, and browse Bear notes directly from your local Bear database. Supports searching by title, content, or tags, opening specific notes, and exploring tag-based organization.

Copilot Studio Agent Direct Line MCP Server

Copilot Studio Agent Direct Line MCP Server

Enables interaction with Microsoft Copilot Studio Agents directly from VS Code through the Direct Line 3.0 API. Supports starting conversations, sending messages, retrieving history, and managing conversation lifecycle with your custom agents.

Tagging MCP

Tagging MCP

Enables parallel tagging and classification of CSV data using multiple LLM providers with structured output, confidence scores, and optional reasoning for batch classification tasks.

mintlify-mcp

mintlify-mcp

An MCP server that enables users to query any Mintlify-powered documentation site directly from Claude. It leverages Mintlify's AI Assistant API to provide RAG-based answers and code examples for various platforms like Agno, Resend, and Upstash.

Algorand MCP Server

Algorand MCP Server

Provides 50+ tools for comprehensive Algorand blockchain development including account management, asset operations, smart contracts, atomic transactions, swap functionality via Pera Swap, and semantic search through Algorand documentation.

MCP Weather Project

MCP Weather Project

An MCP server that provides real-time weather alerts for US states using the National Weather Service (NWS) API. It also includes utility features for message echoing and generating customizable greeting prompts.

tasksync-mcp

tasksync-mcp

MCP server to give new instructions to agent while its working. It uses the get_feedback tool to collect your input from the feedback.md file in the workspace, which is sent back to the agent when you save.

Quarkus Model Context Protocol (MCP) Server

Quarkus Model Context Protocol (MCP) Server

这个扩展程序使开发者能够轻松地实现 MCP 服务器的各项功能。

Test Generator MCP Server

Test Generator MCP Server

Enables automatic generation of test scenarios from user stories uploaded to Claude desktop. Leverages MCP integration to streamline the test case creation process for development workflows.

Local iMessage RAG MCP Server

Local iMessage RAG MCP Server

来自 Anthropic MCP 黑客马拉松 (纽约) 的 iMessage RAG MCP 服务器

Parallels RAS MCP Server (Python)

Parallels RAS MCP Server (Python)

Here are a few possible translations, depending on the nuance you want to convey: **Option 1 (Most Direct):** * **Simplified Chinese:** 使用 FastAPI 的 Parallels RAS 的 MCP 服务器 * **Traditional Chinese:** 使用 FastAPI 的 Parallels RAS 的 MCP 伺服器 This is a straightforward translation. It's clear and accurate. **Option 2 (Slightly More Natural, Emphasizing "Built with"):** * **Simplified Chinese:** 基于 FastAPI 的 Parallels RAS MCP 服务器 * **Traditional Chinese:** 基於 FastAPI 的 Parallels RAS MCP 伺服器 This option uses "基于" (jī yú) which translates to "based on" or "built upon." It implies that the MCP server is constructed using FastAPI. **Option 3 (More Descriptive, Emphasizing "For"):** * **Simplified Chinese:** 用于 Parallels RAS 的基于 FastAPI 的 MCP 服务器 * **Traditional Chinese:** 用於 Parallels RAS 的基於 FastAPI 的 MCP 伺服器 This option uses "用于" (yòng yú) which translates to "for" or "intended for." It emphasizes that the MCP server is designed to be used with Parallels RAS. **Which one should you use?** * If you just need a simple, direct translation, **Option 1** is fine. * If you want to emphasize that the server is *built using* FastAPI, **Option 2** is better. * If you want to emphasize that the server is *designed for* Parallels RAS, **Option 3** is best. In most cases, **Option 1 or 2** will be the most appropriate. Choose the one that best fits the context.

gorse

gorse

Data On Tap Inc. 是一家在加拿大运营网络 302 100 的完整 MVNO(移动虚拟网络运营商)。这是 DOT 的代码仓库。它包括高级安全和身份验证、各种连接工具、智能功能(包括智能网络预留)、eSIM/iSIM、引导无线连接、D2C 卫星、构建框架和概念,以及 OpenAPI 3.1 和 MCP 服务器。