MCP Server Article

🚀 Automatic Article Publishing MCP Server

This project enables the generation of an article using AI and its automatic publication on CSDN, Juejin, and Cnblogs. For details about the MCP protocol, refer to the official MCP documentation.

It is developed relying on the MCP Java SDK, and any intelligent assistant supporting the MCP protocol (such as Claude, Cursor, and Qianfan AppBuilder) can be quickly integrated. More detailed adaptation instructions are provided below.

✨ Features

Tool List

searchExperienceQuestion - Search for Interview Experiences

• Search for interview experiences published by users on Nowcoder.
• Output: Interview experience content

searchNews - Query News

• Query the latest technology news topics from the internet.
• Output: News title, news content

publishArticle2Csdn - Publish Articles to CSDN

• Input: Article title, article content
• Output: Success or failure information of publication

publishArticle2Juejin - Publish Articles to Juejin

• Input: Article title, article content
• Output: Success or failure information of publication

publishArticle2Cnblogs - Publish Articles to Cnblogs

• Input: Article title, article content
• Output: Success or failure information of publication

🔧 Technical Details

Environment Parameters

To configure the MCP server, you need to set the following environment variables:

🚀 Quick Start

📦 Installation

1. Clone the project repository:

   git clone https://github.com/your-repository/mcp-server.git
   cd mcp-server
   

2. Download dependencies and build the project:

   mvn clean install
   

Configuration

Configure the MCP Server

Edit the mcp-servers-config.json file and add the necessary environment variables:

{
  "mcpServers": {
    "articleServer": {
      "command": "java",
      "args": [
        "-Dspring.ai.mcp.server.stdio=true",
        "-Dspring.main.web-application-type=none",
        "-Dlogging.pattern.console=",
        "-jar",
        "/Users/gulihua/IdeaProjects/mcp-server-article/target/mcp-server-0.0.1-SNAPSHOT.jar"
      ],
      "env": {
        "JUEJIN_COOKIE": "Your Juejin Cookie",
        "CNBLOGS_COOKIE": "Your Cnblogs Cookie",
        "CNBLOGS_TOKEN": "Your Cnblogs Token",
        "CSDN_COOKIE": "Your CSDN Cookie"
      }
    }
  }
}

Use Alibaba's Tongyi Qianwen Model

Add the following configuration to the application.yml file:

spring:
  ai:
    dashscope:
      api-key: ${API key of Tongyi Qianwen}
      chat:
        options:
          model: qwen-max

Initialization

Initialize the Chat Client

Inject the ChatClient in the Spring Boot application:

@Bean
public ChatClient initChatClient(ChatClient.Builder chatClientBuilder,
                                 ToolCallbackProvider mcpTools) {
    return chatClientBuilder
            .defaultTools(mcpTools)
            .build();
}

Interface Call

Create a RESTful interface to handle user query requests:

@PostMapping(value = "/ai/answer/sse", produces = MediaType.TEXT_EVENT_STREAM_VALUE)
public Flux<String> generateStreamAsString(@RequestBody AskRequest request) {
    return chatClient.prompt()
            .user(request.getContent())
            .stream()
            .content()
            .concatWith(Flux.just("[complete]"));
}

Through the above steps, you can easily publish AI - generated articles to multiple platforms automatically and support expansion using different intelligent assistants.