发现优秀的 MCP 服务器

KIMP MCP Server

Enables users to query Kimchi Premium (KIMP) data for cryptocurrencies, showing the price difference between Korean and international exchanges for assets like Bitcoin and Ethereum.

Perplexity MCP Server

Okay, I understand. You want to perform a web search using Perplexity to get information for an MCP (presumably a topic you'll provide later), and you want to do this *without* using any API keys. Here's the breakdown of why that's tricky and what options you *might* have, along with their limitations: **Why It's Difficult (and why API keys exist):** * **Perplexity's Design:** Perplexity is designed to be accessed primarily through its official website or its API. The API is the intended way for developers to programmatically access its search and summarization capabilities. APIs are protected by API keys to: * **Control Usage:** Prevent abuse and ensure fair access to resources. * **Track Usage:** Monitor how the service is being used for billing and performance analysis. * **Enforce Terms of Service:** Ensure users adhere to the rules of the platform. * **Web Scraping Challenges:** The alternative to using an API is web scraping (programmatically extracting data from a website). This is generally discouraged and can be problematic: * **Terms of Service Violation:** Most websites (including Perplexity) explicitly prohibit scraping in their terms of service. Violating these terms can lead to your IP address being blocked. * **Website Structure Changes:** Websites change their HTML structure frequently. A scraper that works today might break tomorrow. * **Rate Limiting:** Websites often implement rate limiting to prevent scrapers from overloading their servers. You'll likely be blocked if you make too many requests too quickly. * **Ethical Considerations:** Scraping can put a strain on the website's resources and potentially disrupt service for other users. **Possible (But Limited and Potentially Unreliable) Approaches:** 1. **Manual Search and Copy-Pasting:** * **How it works:** The simplest approach is to manually go to the Perplexity website (perplexity.ai), enter your MCP search query, and then copy and paste the results into your own document or application. * **Pros:** No code required, avoids violating terms of service. * **Cons:** Extremely tedious and time-consuming for anything beyond a few searches. Not automated. 2. **Very Basic Web Scraping (Use with Extreme Caution):** * **Disclaimer:** I strongly advise against this unless you understand the risks and are prepared to deal with potential blocking or legal issues. *Only use this for very small, infrequent, and non-commercial purposes.* * **How it *might* work (but likely won't for long):** * **Inspect the Perplexity Website:** Use your browser's developer tools (usually by pressing F12) to examine the HTML structure of the Perplexity search results page. Identify the HTML elements that contain the search results you want (e.g., `<div>` tags, `<p>` tags, etc.). * **Use a Web Scraping Library:** Use a Python library like `requests` to fetch the HTML content of the Perplexity search results page for your query. Then, use a library like `Beautiful Soup` to parse the HTML and extract the data you identified in the previous step. * **Example (Conceptual - Likely to Break):** ```python import requests from bs4 import BeautifulSoup query = "Your MCP Search Query Here" # Replace with your actual query url = f"https://www.perplexity.ai/search?q={query}" #This is a guess at the URL structure try: response = requests.get(url) response.raise_for_status() # Raise an exception for bad status codes (404, 500, etc.) soup = BeautifulSoup(response.content, "html.parser") # **THIS IS THE TRICKY PART - YOU NEED TO FIND THE RIGHT HTML ELEMENTS** # Example: Let's say the search results are in <div> tags with class "result-item" results = soup.find_all("div", class_="result-item") for result in results: # Extract the text from each result (adjust based on the actual HTML) text = result.text.strip() print(text) except requests.exceptions.RequestException as e: print(f"Error: {e}") except Exception as e: print(f"An unexpected error occurred: {e}") ``` * **Pros:** Potentially automates the search process (but very fragile). * **Cons:** * **High risk of being blocked.** * **Very likely to break due to website changes.** * **Potentially violates Perplexity's terms of service.** * **Requires programming knowledge.** * **No guarantee of accurate or complete results.** **Important Considerations:** * **Ethical Scraping:** If you absolutely must scrape, be respectful: * **Identify Yourself:** Set a `User-Agent` header in your `requests` call to identify your script and provide contact information. * **Rate Limiting:** Introduce delays between requests (e.g., using `time.sleep()`) to avoid overloading the server. * **Check `robots.txt`:** Examine the website's `robots.txt` file (e.g., `perplexity.ai/robots.txt`) to see if there are any specific rules about which pages you are allowed to crawl. * **Consider Alternatives:** Before resorting to scraping, explore other search engines that might offer more accessible APIs or data feeds (even if they aren't exactly Perplexity). **In summary, performing a Perplexity web search without an API key is highly discouraged and comes with significant risks and limitations. The manual approach is the safest, but the least efficient. Web scraping is technically possible, but ethically questionable and practically unreliable.** To give you a more specific answer, please tell me: 1. **What is the MCP you want to search for?** 2. **What is the *purpose* of this search?** (e.g., personal research, academic project, commercial application). This will help me understand the context and suggest more appropriate solutions. 3. **What is your level of programming experience?** Once I have this information, I can provide more tailored advice. However, I must reiterate that I cannot endorse or assist with any activity that violates a website's terms of service.

PromptQL MCP Server

用于 Hasura PromptQL 的模型上下文协议 (MCP) 服务器

MCP Server Hub / Gateway

这个项目提供了一个中心网关来管理多个 MCP（模型上下文协议）服务器，从而避免了为每个 LLM 客户端（如 Cline、Cursor 等）配置和运行重复服务器进程的需求。 将您的 LLM 客户端连接到此项目提供的单个网关客户端端点，即可访问来自所有受管理的 MCP 服务器的工具。

KognitiveKompanion

KDE AI 界面和 MCP 服务器集成

ton-mcp

这个 MCP 服务器允许你与 TON 区块链进行交互。

SimBrief Flight Planning MCP Server

Enables access to SimBrief flight planning data through Claude Desktop with secure Google OAuth authentication. Provides tools to retrieve flight plans, dispatch briefings, NOTAMs, weather data, and other aviation planning information.

MCP: Multi-Agent Control Point

A server that routes user questions to specialized agents (date, location, weather) or an LLM expert, with a simple Streamlit web interface for easy interaction.

GenPilot

GenPilot 简化了由生成式人工智能驱动的多代理系统的创建。它遵循 MCP 协议，并通过直观的终端或 Web 界面确保与 MCP 服务器的顺利集成。

MCP SERVERLY

这是一个旨在表达在官方 Anthropic base SDK for dotnet 9 中模型上下文协议 (MCP) 实现的项目。

Breadcrumb MCP Server

Enables AI assistants to interact with the Breadcrumbs project memory system, allowing creation and management of project documentation including sessions, components, ADRs, patterns, and providing search capabilities across project knowledge.

Sherlog MCP

Provides persistent IPython shell sessions per conversation with DataFrame-centric architecture, enabling stateful data analysis, CLI tool execution, and integration of external MCP servers within the same workspace context.

MCP Reasoning Engine

A production-ready reasoning engine that integrates Claude AI with specialized MCP tools for knowledge retrieval, schema validation, and domain-specific rubric evaluation. It enables structured RAG-based analysis across legal, health, and science domains via a RESTful API.

MCP Google Maps

Provides access to Google Maps API functionality including places search, geocoding, directions, distance matrix, elevation data, and static map generation through the MCP interface.

Databricks MCP Server

Enables AI assistants like Claude to interact with Databricks workspaces through secure OAuth authentication. Supports custom prompts, tools for cluster management, SQL execution, and job operations via the Databricks SDK.

MCP Calculator Demo

A simple demonstration MCP server built with FastMCP that exposes basic calculator operations (add, subtract, multiply, divide) as tools for MCP clients like GitHub Copilot Agent mode.

X (Twitter) MCP server

X (Twitter) MCP server

YNAB MCP Server

An MCP server that provides Large Language Models with access to YNAB (You Need A Budget) budgets, allowing them to fetch budget data including accounts, categories, and category groups.

Octagon Deep Research MCP

Provides specialized AI-powered comprehensive research and analysis capabilities by integrating with advanced deep research agents, offering unlimited queries with no rate limits and faster performance than comparable services.

Amadeus MCP Server

Amadeus MCP (模型上下文协议) 服务器

Anki MCP Server

镜子 (jìng zi)

Edge-TTS MCP Server

A Model Context Protocol server that provides text-to-speech functionality for AI agents using Microsoft Edge's text-to-speech technology, supporting multiple voices, languages, and voice customization.

Hetzner Cloud MCP Server

一个模型上下文协议服务器，允许语言模型通过结构化函数管理 Hetzner Cloud 资源，包括服务器、卷、防火墙和 SSH 密钥。

Contentful

A Model Context Protocol (MCP) server that provides AI assistants with comprehensive tools to interact with Contentful APIs.

Derive MCP

Provides comprehensive access to Lyra Finance's Derive API with 31+ tools for trading options, perpetuals, and spot markets, including real-time market data, historical analytics, account management, and order execution.

Todoist MCP Server

An unofficial MCP server that enables AI agents to create and list tasks in Todoist using natural language. It supports task details such as due dates, priorities, and labels, while allowing for project-based filtering.

Mender MCP Server

Enables AI assistants to interact with Mender IoT platform for device management, deployment monitoring, and fleet analysis through natural language commands. Provides read-only access to device status, deployment logs, releases, and system monitoring capabilities.

Tessie MCP Extension

Enables Claude Desktop to access Tesla vehicle data through the Tessie API. Users can query their car's location, battery level, mileage, driving history, and charging status using natural language.