Weiwanmcpserver

🚀 Show Time: AI Geek Marathon Project - Super Chat + Custom MCP Server

This project is a program based on Super Chat (an open - source chat client) and a custom MCP server developed for the AI Geek Marathon competition. It enables users to call MCP tools through chat interactions, achieving the following core functions:

1. Network Information Retrieval: Search for materials related to keywords on specified websites (e.g., Zhihu) and return results in the form of links + summaries.
2. Note Download: Download the content of retrieved links as Markdown notes and save them to the local note repository.
3. Note Management: Perform operations such as viewing, summarizing, and deleting notes (based on the open - source implementation of Obsidian).
4. Note Rewriting: Rewrite or organize the content of downloaded notes.

🚀 Quick Start

The project combines Super Chat and a custom MCP server. Users can interact with the system through Super Chat to call various MCP tools for information retrieval, note download, and management.

✨ Features

1. Seamless Integration: Natural language interaction via Super Chat reduces the user's learning curve.
2. Efficient Retrieval: Supports retrieval on specified websites, enabling quick access to target information.
3. Localized Management: Saves online content as Markdown notes for easy subsequent organization and reuse.
4. Flexibility: Supports note rewriting and integration with Obsidian to meet personalized needs.

📦 Installation

There is no specific installation process provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

Users can use Super Chat to call MCP tools. For example, they can use the search_news tool to retrieve network information, the download_note tool to download web content as notes, and the overwrite_note tool to rewrite or organize note content.

Advanced Usage

The system also supports more advanced operations, such as specifying the website for information retrieval and managing local notes.

# MCP server core code
from mcp.server.fastmcp import FastMCP
from datetime import datetime
import os
import requests
import re

mcp = FastMCP("WeiWanMcp", instructions="""
    Restrictions:
        1. After calling search_news to query information, do not summarize the content and output it directly.
""")

nodePath = os.getenv("NOTES_PATH", "E:\MyDocuments\MyNotes\Clipper")

@mcp.tool()
def search_news(query: str, website: str):
    """Retrieve network information"""
    url = 'https://s.jina.ai/?q=' + query + '&num=20'
    headers = {
        'Authorization': 'Bearer YOUR_API_KEY',
        'X-Respond-With': 'no-content',
    }
    if website:
        if validate_domain(website):
            headers['X-Site'] = website
    response = requests.get(url, headers=headers)
    return response.text

@mcp.tool()
def download_note(webLink: str, nodeTitle: str):
    """Download web content as a note"""
    headers = {"Authorization": "Bearer YOUR_API_KEY"}
    response = requests.get("https://r.jina.ai/" + webLink, headers=headers)
    with open(nodePath + "\\" + nodeTitle + ".md", "w", encoding='utf-8') as file:
        file.write(response.text)
    return "Note saved to: " + nodePath + "\\" + nodeTitle + ".md"

@mcp.tool()
def overwrite_note(webLink: str, nodeTitle: str):
    """Rewrite or organize note content"""
    headers = {"Authorization": "Bearer YOUR_API_KEY"}
    response = requests.get("https://r.jina.ai/" + webLink, headers=headers)
    with open(nodePath + "\\" + nodeTitle + ".md", "w", encoding='utf-8') as file:
        file.write(response.text)
    return "Note rewritten to: " + nodePath + "\\" + nodeTitle + ".md"

📚 Documentation

1. Core Components

• Super Chat: Serves as the front - end for user interaction, supporting chat - style calls to MCP tools.
• Custom MCP Server: Implemented in Python, providing the following tools:
  • search_news: Retrieve network information.
  • download_note: Download web content as Markdown notes.
  • overwrite_note: Rewrite or organize note content.
  • Other auxiliary tools (e.g., getting the current time, date, etc.).

2. Code Highlights

The core code of the MCP server demonstrates how to implement functions such as information retrieval, note download, and rewriting. It uses the requests library for HTTP requests, ensuring efficient and stable data acquisition.

3. Implementation Details

• The server - side code implements functions for network information retrieval, download, and rewriting, supporting precise searches on specified websites (e.g., Zhihu).
• The requests library is used for HTTP request processing to ensure efficient and stable data acquisition.
• Note content is stored in local Markdown files for easy subsequent editing and management.

🔧 Technical Details

The project uses Python to implement the custom MCP server. The FastMCP class is used to manage tools, and the requests library is used for HTTP requests. The core code defines several key tools, including search_news, download_note, and overwrite_note, which are used to achieve information retrieval, note download, and rewriting.

📄 License

No license information is provided in the original document, so this section is skipped.

📷 Function Screenshots

🌟 Conclusion

This project combines Super Chat and a custom MCP server to achieve full - process automation from information retrieval to note management. In the future, the functions will be further optimized to provide users with more powerful knowledge management tools!