Documentation Hub MCP Server

🚀 Document Knowledge Base MCP Server

This is a Model Context Protocol (MCP) server that provides intelligent document retrieval across multiple technical domains.

🚀 Quick Start

The project implements an MCP server for targeted document retrieval in response to technical questions. It uses a hybrid approach to:

1. Determine the technical domain of the query (React, Node.js, Python).
2. Extract specific topics from the query.
3. Construct precise document URLs.
4. Fetch and return the most relevant documents.

The server is built with TypeScript and the MCP SDK, providing a standardized interface for AI models to access document resources.

✨ Features

• Domain Classification: Automatically identify whether a query belongs to React, Node.js, Python, or a general topic.
• Topic Extraction: Use regular expression pattern matching to identify specific technical concepts in the query.
• URL Construction: Build target document URLs based on the identified topics.
• Multi - Document Source Support: Support documentation for React, Node.js, and Python, and can be extended to other domains.
• Standardized Interface: Implement the Model Context Protocol for interoperability with MCP - compatible clients.

📦 Installation

# Clone the repository
git clone https://github.com/yourusername/mcp-server.git
cd mcp-server

# Install dependencies
npm install

# Build the project
npm run build

💻 Usage Examples

Basic Usage

Run the Server

node build/index.js

Test Topic Extraction

The project includes a test script that demonstrates how topic extraction and URL construction work:

# Compile the test script
npx tsc src/test.ts --outDir build

# Run the test
node build/test.js

Integrate with a Client

The server can be integrated with any MCP - compatible client. Here is an example client code (not in this project):

// Example client code
const client = new McpClient();
const result = await client.callTool("fetch-documentation", {
  query: "How to use the useState hook in React?",
});

console.log(result);
// Output:
// {
//   domain: "react-docs",
//   topics: ["useState", "state"],
//   specificUrl: "https://react.dev/reference/react/useState",
//   content: "...",
//   source: "https://react.dev/reference/react/useState"
// }

📚 Documentation

Project Structure

• src/index.ts - Main server implementation.
• src/test.ts - Test script for topic extraction and URL construction.
• build/ - Compiled JavaScript output.
• package.json - Project configuration and dependencies.

Available Tools

determine - domain

Determine which technical domain a query belongs to.

{
  query: string; // User query to be classified
}

Returns:

{
  domain: string; // Identified domain
  confidence: string; // "high" or "low"
}

extract - topics

Extract specific topics from a query based on a given domain.

{
  query: string; // User query to be analyzed
  domain: string; // Technical domain to extract topics from
}

Returns:

{
  topics: string[]; // Array of identified topics
  count: number; // Number of topics found
}

fetch - documentation

Fetch documentation based on a query and an optional domain.

{
  query: string; // User query content
  domain?: string; // Domain (optional)
}

Returns:

{
  "docs": [
    {
      "url": string,
      "content": string,
      "relevance": number
    }
  ],
  "timestamp": string,
  "status": "success" | "error"
}

Available Service Resources

• Document Storage: Used for storing and retrieving technical documents.
• Topic Index: Used for quickly finding documents related to specific topics.
• URL Builder: Constructs target URLs based on queries.

Expansion

Add a New Domain

To add a new technical domain (e.g., a new document storage), follow these steps:

1. Create a new collection or table in the database.
2. Update the extract - topics tool to recognize topics in the new domain.
3. Test the new feature to ensure it works properly.

Improve Topic Extraction

To improve the accuracy of topic extraction, consider the following methods:

• Use more complex natural language processing models (e.g., BERT).
• Increase the amount of training data.
• Regularly update and optimize regular expression patterns.

📄 License

[Insert license content here]

Contribution Guidelines

Contributions to this project are welcome! Please refer to the CONTRIBUTING.md file in the project repository to learn how to participate.