发现优秀的 MCP 服务器

LlamaCloud MCP Server

一个本地 MCP 服务器，与 Claude Desktop 集成，启用 RAG 功能，使 Claude 能够从自定义 LlamaCloud 索引获取最新的私有信息。

callput-agent-mcp

MCP server for spread-only options execution on Base L2, enabling AI agents to trade options via a defined tool loop.

AI Consultant MCP Server

Enables AI agents to consult with multiple AI models (GPT, Gemini, Grok, etc.) through OpenRouter with intelligent auto-selection, conversation history, and caching. Allows your AI assistant to seek expert opinions from specialized models for different tasks like coding, analysis, or general questions.

TaskMCP

A personal task management MCP server that allows LLM clients to create, read, update, and delete tasks with projects, labels, and comments, using a local SQLite database that can also be accessed via a web UI.

astrology-mcp

MCP server for astrological calculations including natal charts, planetary positions, aspects, transits, and progressions using Swiss Ephemeris.

LocalDB MCP Server

Enables AI assistants to query and interact with Microsoft SQL Server LocalDB databases, supporting operations like listing databases, querying tables, executing SQL statements, and viewing stored procedures.

CodeGraph

CodeGraph — Open-source code intelligence MCP server. Builds a semantic graph of your codebase (functions, classes, imports, call chains) and exposes it through 31 tools. Callers, callees, impact analysis, complexity metrics, unused code detection, AI context assembly, persistent memory, cross-project search. 15 languages via tree-sitter. Single Rust binary, local-first.

PromtPress Ads MCP Server

MCP server for buying advertising outcomes such as Telegram subscribers and website traffic from a prepaid balance. Provides tools for campaign management, forecasting, and budget control, with guardrails so agents can spend but not fund or withdraw.

KaiaFun MCP

一个MCP服务器，通过Claude Desktop实现代币上架、交易以及与Kaia区块链的交互。

Bolsai

Provides Brazilian financial market data including stock prices, fundamentals, dividends, FIIs, and macro indicators to AI assistants.

Armorial

Enables AI agents and humans to resolve, search, and retrieve deterministic, policy-compliant SVG icons from the IconPark library, with explicit ambiguity handling and visual decision support.

MCP Demo

Okay, I can't directly "demonstrate an MCP server" in the sense of running code and showing you output here. That requires a real server environment and access to aviation weather data APIs. However, I can provide you with a conceptual outline and code snippets (in Python, a common language for this) to illustrate how such a server *could* be built. This will give you a solid understanding of the components involved. **Conceptual Outline** 1. **Data Source:** The server needs to fetch aviation weather data from a reliable source. Common sources include: * **NOAA Aviation Weather Center (AWC):** Provides METARs, TAFs, PIREPs, and other aviation weather products. Often accessed via their XML/text feeds. * **Aviation Weather APIs (Commercial):** Some companies offer paid APIs that provide more structured data and potentially better performance. Examples include Aviation Edge, CheckWX, etc. 2. **Server Framework:** Choose a web server framework to handle incoming requests and send responses. Popular choices include: * **Flask (Python):** Lightweight and easy to learn. Good for simple APIs. * **FastAPI (Python):** Modern, high-performance, and automatically generates API documentation. * **Node.js (JavaScript):** If you prefer JavaScript. Express.js is a common framework. 3. **Data Parsing and Storage (Optional):** * **Parsing:** The data from the source (e.g., XML from NOAA) needs to be parsed into a usable format (e.g., Python dictionaries or objects). * **Storage (Optional):** For performance, you might want to cache the weather data in a database (e.g., Redis, PostgreSQL) or in memory. This avoids hitting the external API too frequently. 4. **API Endpoints:** Define the API endpoints that clients will use to request data. For example: * `/metar/{icao}`: Get the METAR for a specific airport (ICAO code). * `/taf/{icao}`: Get the TAF for a specific airport. * `/airports/search?q={query}`: Search for airports by name or ICAO code. 5. **Error Handling:** Implement proper error handling to gracefully handle issues like invalid airport codes, API errors, and network problems. 6. **Security (Important):** If the server is publicly accessible, implement security measures to prevent abuse. This might include rate limiting, authentication, and authorization. **Python (Flask) Example Code Snippets** ```python from flask import Flask, jsonify, request import requests import xml.etree.ElementTree as ET # For parsing XML (if using NOAA) app = Flask(__name__) # Replace with your actual NOAA ADDS URL or other API endpoint NOAA_ADDS_URL = "https://aviationweather.gov/adds/dataserver/.......your_query_here......." # Example, needs a real query # In-memory cache (for demonstration purposes only; use a real database for production) weather_cache = {} def fetch_metar_from_noaa(icao): """Fetches METAR data from NOAA ADDS for a given ICAO code.""" try: # Construct the NOAA ADDS query (example, adjust as needed) query = f"?dataSource=metars&requestType=retrieve&format=xml&stationString={icao}&hoursBeforeNow=1" url = NOAA_ADDS_URL.replace(".......your_query_here.......", query) response = requests.get(url) response.raise_for_status() # Raise HTTPError for bad responses (4xx or 5xx) xml_data = response.text root = ET.fromstring(xml_data) # Parse the XML to extract the METAR information metar_element = root.find(".//METAR/raw_text") # Adjust path based on XML structure if metar_element is not None: metar_text = metar_element.text return metar_text else: return None # METAR not found except requests.exceptions.RequestException as e: print(f"Error fetching data from NOAA: {e}") return None except ET.ParseError as e: print(f"Error parsing XML: {e}") return None @app.route('/metar/<icao>') def get_metar(icao): """API endpoint to get METAR data for a given ICAO code.""" icao = icao.upper() # Ensure ICAO is uppercase # Check the cache first if icao in weather_cache: print(f"Fetching METAR for {icao} from cache") metar = weather_cache[icao] else: print(f"Fetching METAR for {icao} from NOAA") metar = fetch_metar_from_noaa(icao) if metar: weather_cache[icao] = metar # Store in cache if metar: return jsonify({'icao': icao, 'metar': metar}) else: return jsonify({'error': 'METAR not found for this ICAO code'}), 404 @app.route('/airports/search') def search_airports(): """Example endpoint for searching airports (replace with real implementation).""" query = request.args.get('q') if not query: return jsonify({'error': 'Missing query parameter'}), 400 # Replace this with a real airport search implementation (e.g., from a database) # This is just a placeholder if query.lower() == "jfk": results = [{"icao": "KJFK", "name": "John F. Kennedy International Airport"}] elif query.lower() == "lax": results = [{"icao": "KLAX", "name": "Los Angeles International Airport"}] else: results = [] return jsonify(results) if __name__ == '__main__': # Production: Use a proper WSGI server (e.g., gunicorn, uWSGI) app.run(debug=True) # Debug mode for development only ``` **Explanation of the Code:** * **Imports:** Imports necessary libraries (Flask, `requests` for making HTTP requests, `xml.etree.ElementTree` for parsing XML). * **`NOAA_ADDS_URL`:** **CRITICAL:** You *must* replace this with a valid URL for the NOAA ADDS server. The example is just a placeholder. You'll need to construct the correct query parameters to get the data you want. Refer to the NOAA ADDS documentation. * **`weather_cache`:** A simple in-memory dictionary to cache weather data. This is for demonstration only. In a real application, use a database like Redis. * **`fetch_metar_from_noaa(icao)`:** * Constructs the NOAA ADDS URL with the ICAO code. * Uses `requests` to fetch the data. * Parses the XML response using `xml.etree.ElementTree`. * Extracts the METAR text from the XML. **Important:** The XML structure can be complex. You'll need to carefully inspect the NOAA ADDS XML response to determine the correct XPath to the METAR data. * Handles potential errors (network errors, XML parsing errors). * **`get_metar(icao)`:** * The API endpoint for `/metar/{icao}`. * Checks the cache first. * If not in the cache, fetches the data from NOAA. * Returns the METAR data as a JSON response. * Returns a 404 error if the METAR is not found. * **`search_airports()`:** A placeholder for an airport search endpoint. You'll need to replace this with a real implementation that queries a database of airports. * **`app.run(debug=True)`:** Starts the Flask development server. **Important:** Do not use `debug=True` in a production environment. Use a proper WSGI server (e.g., gunicorn, uWSGI). **To Run This Example (After Replacing the Placeholder URL):** 1. **Install Flask and Requests:** ```bash pip install flask requests ``` 2. **Save the code:** Save the code as a Python file (e.g., `aviation_server.py`). 3. **Run the server:** ```bash python aviation_server.py ``` 4. **Test the API:** Open a web browser or use `curl` to test the API endpoints: * `http://127.0.0.1:5000/metar/KJFK` (Replace `KJFK` with a valid ICAO code) * `http://127.0.0.1:5000/airports/search?q=jfk` **Key Improvements and Considerations for a Production System:** * **Error Handling:** More robust error handling, including logging and more informative error messages. * **Configuration:** Use environment variables or a configuration file to store API keys, database connection strings, and other settings. * **Data Validation:** Validate the ICAO code and other input parameters to prevent errors and security vulnerabilities. * **Rate Limiting:** Implement rate limiting to prevent abuse of the API. * **Authentication/Authorization:** If the API is sensitive, implement authentication and authorization to control access. * **Asynchronous Operations:** For better performance, use asynchronous operations (e.g., `asyncio` in Python) to fetch data from the external API without blocking the server. * **Testing:** Write unit tests and integration tests to ensure the server is working correctly. * **Deployment:** Deploy the server to a production environment (e.g., AWS, Google Cloud, Azure) using a proper WSGI server (e.g., gunicorn, uWSGI). * **Monitoring:** Monitor the server's performance and error rates. * **TAF Support:** Implement the `/taf/{icao}` endpoint to fetch Terminal Aerodrome Forecasts. This will involve a similar process of querying the NOAA ADDS server (or another API) and parsing the TAF data. * **Data Source Abstraction:** Create an abstraction layer for the data source. This will make it easier to switch to a different API in the future. **Chinese Translation of Key Concepts** * **MCP Server:** MCP服务器 (MCP fúwùqì) - While "MCP" isn't a standard term in this context, it's understood as a server providing specific data. A more descriptive term might be 航空气象数据服务器 (Hángkōng qìxiàng shùjù fúwùqì) - Aviation Weather Data Server. * **Aviation Weather Data:** 航空气象数据 (Hángkōng qìxiàng shùjù) * **METAR:** 机场气象报告 (Jīchǎng qìxiàng bàogào) * **TAF:** 机场预报 (Jīchǎng yùbào) * **ICAO Code:** 国际民航组织机场代码 (Guójì Mínháng Zǔzhī jīchǎng dàimǎ) * **API Endpoint:** API端点 (API duāndiǎn) * **Data Source:** 数据源 (Shùjù yuán) * **Parsing:** 解析 (Jiěxī) * **Caching:** 缓存 (Huǎncún) * **Error Handling:** 错误处理 (Cuòwù chǔlǐ) * **Rate Limiting:** 速率限制 (Sùlǜ xiànzhì) * **Authentication:** 身份验证 (Shēnfèn yànzhèng) * **Authorization:** 授权 (Shòuquán) This comprehensive explanation and code example should give you a strong foundation for building your own aviation weather data server. Remember to replace the placeholder URL with a valid NOAA ADDS query and adapt the code to your specific needs. Good luck!

mcp

Provides MCP servers for GitHub API operations and SQL database queries, enabling users to interact with GitHub repositories and databases through natural language.

worldcup-mcp

MCP server exposing live 2026 FIFA World Cup group standings and Round-of-32 bracket, with confidence classification and stale fallback.

Data.gov MCP Server

镜子 (jìng zi)

ntfy-mcp

Enables sending push notifications via ntfy with a single tool, allowing Claude agents to send notifications directly without shell access.

website-intelligence-provider

Enables deterministic bilingual website intelligence audits by URL, returning fixed English or Spanish findings without ever contacting the submitted site.

YouTube MCP Server

一个服务器，它通过模型上下文协议实现与 YouTube 数据的交互，允许用户搜索视频、检索视频/频道的详细信息以及获取评论。

Scrapy MCP Server

A powerful web scraping MCP server built on Scrapy and FastMCP that supports multiple scraping methods (HTTP, Scrapy, browser automation), anti-detection techniques, form handling, and concurrent crawling. Designed for commercial environments with enterprise-grade features like intelligent retry mechanisms, performance monitoring, and configurable data extraction.

eanscan-mcp

MCP server for looking up EAN/UPC barcodes to get localized product details like title, description, price, and images.

@a-bonus/redmine-mcp

MCP server for Redmine that enables interaction with issues, wiki pages, and time entries through natural language.

ai-trader-station

MCP server exposing Bill Williams trading signal analysis, market data, and order execution capabilities from MetaTrader 5, enabling AI assistants to run analyses and execute trades through natural language.

llm-fallback-mcp

MCP server that completes prompts with automatic provider fallback across OpenAI, Anthropic Claude, and Google Gemini. It is rate-limit aware, retries on transient failures, and provides a transparent attempt log.

AJEMS MCP Connector

Enables AI assistants to query and optionally modify AJEMS workspaces, including apps, forms, and records, without requiring user login.

ansible-know-mcp

Module discovery, documentation search, and skill generation for AI agents via the Model Context Protocol.

PHP MCP Protocol Server

MCP 通用 PHP 服务器 - 将 PHP 与模型上下文协议集成

smart-fhir

Enables read-only SMART on FHIR R4 discovery, search, and read operations against public synthetic sandboxes, returning actual empty results and 404s rather than inventing patient data.

LinkedIn Custom MCP Server

Enables AI agents to manage professional networking on LinkedIn by providing tools for posting updates, searching for jobs, and analyzing profiles. It facilitates secure interaction with the LinkedIn platform through OAuth 2.0 authentication and the Model Context Protocol.

n8n MCP Server

Enables full workflow automation management in n8n through 40+ tools covering workflows, executions, credentials, tags, variables, projects, users, and source control operations.

fagents-mcp

MCP server that enables Claude Code agents to send and read emails via SMTP/IMAP with per-agent credential isolation and audit logging.