发现优秀的 MCP 服务器

mcp-server-macos-use

能够使用操作系统级别工具控制计算机的人工智能代理，兼容 MCP，并且可以与任何模型一起使用。

browser-for-ai

A CDP-native MCP server that lets AI agents drive a real Chrome browser with deep network/console inspection, robust page interaction, and the ability to reverse-engineer a site's API flow into runnable code with dependency chaining and replay verification.

Claude Talk to Figma MCP

Enables AI assistants like Claude and Cursor to interact directly with Figma to create, modify, and analyze design elements in real-time. It provides a comprehensive suite of tools for document inspection, styling, component management, and automated layout via a bidirectional WebSocket connection.

bexio-mcp

A standalone MCP server for the bexio REST API, enabling interaction with bexio resources like contacts, invoices, and projects through natural language from any MCP client.

@bolagsapi/mcp-server

MCP server for Swedish company data. Enables AI agents to lookup companies, analyze financials, assess health, screen compliance, and get industry stats.

Nectar-MCP

Nectar is a stdio MCP server for Pollinations image and video generation. It gives MCP clients a small, focused toolset for generating images, editing images, generating videos, listing available image/video models, and checking pollen balance.

Ghibli Image Generator MCP Server

Provides access to the Ghibli Image Generator API for creating Ghibli-style images using OpenAI models. It enables users to generate stylized artwork through the ghibli_generate_image tool.

Custom Search MCP Server

An MCP server that enables interaction with Google Custom Search API through natural language, allowing users to perform searches programmatically via the Multi-Agent Conversation Protocol.

devto-mcp-server

Provides tools to search and retrieve articles from the dev.to API, including search by query, tag, or technology, and fetching full articles by ID or slug.

Kali SSE MCP Command Executor

Enables secure execution of penetration testing commands on Kali Linux through Server-Sent Events with intelligent command validation, real-time monitoring, and comprehensive audit logging. Designed for authorized security research and penetration testing workflows.

GitEquity MCP Server

MCP server that lets AI assistants read portfolios and propose trades on GitEquity, with all writes requiring a GitHub-authorized confirmation code for safety.

Spark Customer Agent MCP Server

Enables natural language shopping through Walmart's backend API, supporting product discovery, cart management, coupon handling, and order history.

Wise MCP Server

Read-only access to Wise (TransferWise) personal API for profiles, balances, exchange rates, transfers, and recipients.

weather-mcp-server

Enables Claude Desktop to query real-time weather information, forecasts, and history via the OpenWeatherMap API.

Employee MCP Server (Mongo + AI Pipeline)

Enables querying administration-style employee data in MongoDB using natural language through various AI providers.

VNStock MCP Server

Okay, here's a breakdown of how you could create an MCP (presumably meaning a Minimal, Complete, and Verifiable example) server in Python to fetch historical stock prices using the `vnstock` library, along with explanations and considerations: ```python # server.py (or whatever you want to name your server file) from flask import Flask, request, jsonify import vnstock import datetime app = Flask(__name__) @app.route('/historical_stock_data', methods=['GET']) def get_historical_data(): """ Fetches historical stock data for a given ticker symbol and date range using the vnstock library. Query Parameters: ticker (str): The stock ticker symbol (e.g., 'VIC'). Required. start_date (str): The start date in 'YYYY-MM-DD' format. Required. end_date (str): The end date in 'YYYY-MM-DD' format. Defaults to today if not provided. Returns: JSON: A JSON response containing the historical stock data as a list of dictionaries, or an error message if there's an issue. """ ticker = request.args.get('ticker') start_date = request.args.get('start_date') end_date = request.args.get('end_date') if not ticker: return jsonify({'error': 'Missing ticker parameter'}), 400 # Bad Request if not start_date: return jsonify({'error': 'Missing start_date parameter'}), 400 try: datetime.datetime.strptime(start_date, '%Y-%m-%d') # Validate start_date format if end_date: datetime.datetime.strptime(end_date, '%Y-%m-%d') # Validate end_date format else: end_date = datetime.date.today().strftime('%Y-%m-%d') # Default to today except ValueError: return jsonify({'error': 'Invalid date format. Use YYYY-MM-DD.'}), 400 try: data = vnstock.stock_historical_data( symbol=ticker, start_date=start_date, end_date=end_date, period='1D' # Daily data ) # Convert DataFrame to list of dictionaries for JSON serialization data_list = data.to_dict(orient='records') return jsonify(data_list), 200 # OK except Exception as e: print(f"Error fetching data: {e}") # Log the error for debugging return jsonify({'error': f'Error fetching data: {str(e)}'}), 500 # Internal Server Error if __name__ == '__main__': app.run(debug=True, host='0.0.0.0', port=5000) # Important for network access ``` **Explanation and Key Improvements:** 1. **Dependencies:** - Make sure you have the necessary libraries installed: ```bash pip install flask vnstock ``` 2. **Flask Setup:** - `Flask` is a lightweight Python web framework. We create a Flask app instance. 3. **Route Definition (`/historical_stock_data`):** - `@app.route('/historical_stock_data', methods=['GET'])`: This defines a route that listens for GET requests at the `/historical_stock_data` endpoint. GET is appropriate for fetching data. 4. **Query Parameters:** - `request.args.get('ticker')`: This retrieves the `ticker`, `start_date`, and `end_date` from the URL's query parameters. For example: - `http://localhost:5000/historical_stock_data?ticker=VIC&start_date=2023-01-01&end_date=2023-01-10` 5. **Input Validation:** - **Required Parameters:** Checks if `ticker` and `start_date` are provided. Returns a 400 error (Bad Request) if they are missing. - **Date Format Validation:** Uses `datetime.datetime.strptime` to validate that `start_date` and `end_date` are in the correct `YYYY-MM-DD` format. Returns a 400 error if the format is invalid. - **Default `end_date`:** If `end_date` is not provided, it defaults to the current date. 6. **`vnstock.stock_historical_data()` Call:** - `vnstock.stock_historical_data(...)`: This is where the `vnstock` library is used to fetch the historical data. The `symbol`, `start_date`, and `end_date` are passed as arguments. `period='1D'` specifies daily data. 7. **Error Handling:** - `try...except`: A `try...except` block is used to catch potential errors during the `vnstock` data fetching process. This is crucial for a robust server. - **Logging:** `print(f"Error fetching data: {e}")` logs the error to the console. This is very helpful for debugging. In a production environment, you'd want to use a more sophisticated logging system. - **Error Response:** If an error occurs, a JSON response with an error message and a 500 status code (Internal Server Error) is returned. 8. **JSON Response:** - `data.to_dict(orient='records')`: Converts the Pandas DataFrame returned by `vnstock` into a list of dictionaries. This is necessary because Flask's `jsonify` function can easily serialize lists of dictionaries into JSON. - `jsonify(data_list)`: Converts the list of dictionaries into a JSON response. - `return jsonify(data_list), 200`: Returns the JSON response with a 200 status code (OK). 9. **Running the App:** - `if __name__ == '__main__':`: This ensures that the app is only run when the script is executed directly (not when it's imported as a module). - `app.run(debug=True, host='0.0.0.0', port=5000)`: - `debug=True`: Enables debug mode, which provides helpful error messages and automatic reloading when you make changes to the code. **Important:** Disable debug mode in production. - `host='0.0.0.0'`: This makes the server accessible from any IP address on your network. If you only want to access it from your local machine, use `host='127.0.0.1'`. - `port=5000`: Specifies the port number the server will listen on. **How to Run:** 1. **Save:** Save the code as `server.py` (or any name you prefer). 2. **Install:** Make sure you have Flask and vnstock installed (`pip install flask vnstock`). 3. **Run:** Open a terminal or command prompt, navigate to the directory where you saved the file, and run: `python server.py` 4. **Access:** Open a web browser or use a tool like `curl` or `Postman` to access the server. For example: ``` http://localhost:5000/historical_stock_data?ticker=VIC&start_date=2023-01-01&end_date=2023-01-10 ``` **Important Considerations:** * **Error Handling:** The error handling in this example is basic. In a production environment, you'd want to implement more robust error handling, including logging to a file, sending error notifications, and potentially retrying failed requests. * **Security:** This is a very basic example and doesn't include any security measures. If you're deploying this to a public server, you'll need to consider security aspects like authentication, authorization, and input validation to prevent malicious attacks. * **Rate Limiting:** Be mindful of the API usage limits of the `vnstock` library or the underlying data source. Implement rate limiting in your server to avoid being blocked. * **Asynchronous Operations:** For handling multiple concurrent requests efficiently, consider using asynchronous frameworks like `asyncio` and `aiohttp` instead of Flask. * **Configuration:** Use environment variables or a configuration file to store sensitive information like API keys or database credentials. * **Deployment:** Consider using a production-ready web server like Gunicorn or uWSGI to deploy your Flask application. **Chinese Translation of Key Terms:** * **Stock Price:** 股票价格 (gǔpiào jiàgé) * **Historical Data:** 历史数据 (lìshǐ shùjù) * **Ticker Symbol:** 股票代码 (gǔpiào dàimǎ) * **Start Date:** 开始日期 (kāishǐ rìqí) * **End Date:** 结束日期 (jiéshù rìqí) * **Server:** 服务器 (fúwùqì) * **API:** 应用程序接口 (yìngyòng chéngxù jiēkǒu) * **JSON:** JSON 数据格式 (JSON shùjù géshì) * **Error:** 错误 (cuòwù) * **Request:** 请求 (qǐngqiú) * **Response:** 响应 (xiǎngyìng) This comprehensive example should give you a solid foundation for building your stock price API using `vnstock` and Flask. Remember to adapt it to your specific needs and consider the important considerations mentioned above.

Msty Admin MCP

Comprehensive MCP server for administering Msty Studio Desktop with 36 tools across 6 phases, Bloom behavioral evaluation, and support for four service backends.

Daily Briefing MCP Server

Aggregates data from Google Calendar, TripIt, and Fireflies.ai to generate comprehensive daily briefings, schedule overviews, and action item lists. It enables users to manage their time by detecting meeting conflicts, identifying focus slots, and tracking upcoming travel plans.

mcp-realestate

Provides access to actual residential and commercial real estate transaction prices in Japan.

Odoo MCP Server

Exposes Odoo 19 ERP as MCP tools, enabling AI agents to query and act on CRM data (leads, semantic search) via natural language without custom integration.

unofficial-polestar-mcp

MCP server for controlling Polestar vehicles via natural language, enabling climate, charging, locks, location, and vehicle status management through Claude.

DALL-E MCP Server

An MCP server that provides tools to generate, edit, and create variations of images using OpenAI DALL-E models.

MCP LocalStack Server

Provides conversational access to LocalStack AWS services (S3, DynamoDB, Lambda, SQS) through natural language, enabling local cloud infrastructure management without real AWS costs.

Git MCP Server

Enables git repository operations through REST endpoints, providing access to repository status, diffs, and commits. Enforces security through configurable root directory allowlists for safe git operations.

Jira MCP Server

Enables AI assistants to interact with Atlassian Jira Cloud, allowing users to manage projects, issues, comments, and workflows through natural language commands.

federal-spend-ai

MCP server for analyzing Canadian federal spending data, offering tools for contract search, NLP, semantic search, anomaly detection, and money-flow tracing.

cambium-remote

Enables recall of promoted team and org knowledge from GitHub repositories via Claude.ai, supporting search and status checks.

mcp-compressor

A proxy that intercepts MCP responses to reduce token consumption by compressing them via a pipeline and exposing only two meta-tools to the LLM.

Hetzner MCP Server

Enables management of Hetzner Cloud resources including servers, SSH keys, volumes, storage boxes, and live metrics through 40 tools.

FastMCP Runtime MCP

Paid remote MCP for hosted MCP server providing structured receipts, usage logs, and audit-ready evidence for agent and CI workflows.