Websearch (Google)

🚀 Web Search

WebSearch is a tool for web information retrieval that supports interaction with models via the MCP protocol. It can parse structured search results and offers various customization options to meet the needs of different scenarios.

🚀 Quick Start

WebSearch can interact with models through the MCP protocol. Here is a simple usage example:

import { createMCPClient } from '@modelcontextprotocol/sdk';

// Create an MCP client
const client = createMCPClient({
  transport: { type: 'subprocess', command: 'npx websearch-mcp' }
});

// Perform a web search
const response = await client.request({
  method: 'call_tool',
  params: {
    name: 'web_search',
    arguments: {
      query: 'machine learning trends',
      numResults: 5,
      language: 'en'
    }
  }
});

console.log(response.result);

✨ Features

Core Features

• Web Search: Supports information retrieval through multiple search engines.
• Result Parsing: Can parse structured search results, including titles, summaries, links, etc.
• Multi-language Support: Supports returning search results in multiple languages.

Customization Options

• Number of Results: You can specify the number of results to return, with a default of 5.
• Language Filtering: Specify the language of the search results.
• Domain Exclusion/Inclusion: You can selectively exclude or include results from specific domains.
• Keyword Filtering: Supports excluding specific keywords or phrases.

Example Response

Here is a typical example of a search response:

{
  "query": "machine learning trends",
  "results": [
    {
      "title": "2025 AI Trend Predictions",
      "snippet": "The development trends of AI in 2025 include the application of multi-modal AI, generative models, and quantum AI in enterprises...",
      "url": "https://example.com/ml-trends-2025",
      "siteName": "AI Research Today",
      "byline": "Dr. Jane Smith"
    },
    {
      "title": "The Evolution of AI from 2020 to 2025",
      "snippet": "In the past five years, AI has evolved from supervised learning methods to more complex self-supervised and reinforcement learning paradigms...",
      "url": "https://example.com/ml-evolution",
      "siteName": "Tech Insights",
      "byline": "John Doe"
    }
  ]
}

📦 Installation

Install as a Dependency

npm install websearch-mcp-sdk

Clone the Repository

git clone https://github.com/yourusername/websearch.git
cd websearch
npm install

📚 Documentation

Development Environment Requirements

• Operating System: Supports Linux, Windows, and macOS.
• Node.js Version: Requires Node.js 14 or higher.

Project Structure

websearch/
├── .gitignore
├── .npmignore
├── package.json
├── src/
│   ├── index.ts
│   └── types/
└── dist/
    └── main.js

Development Steps

1. Clone the repository to your local machine.
2. Install project dependencies: npm install
3. Start the development server: npm run dev

🔧 Troubleshooting

Common Issues

• API Unreachable: Ensure that the WebSearch service is running and the API_URL is configured correctly.
• Search Results Unavailable: Check the logs of the WebSearch service to rule out errors:

  docker-compose logs crawler
  

• FlareSolverr Issues: If you encounter Cloudflare protection, check if FlareSolverr is running properly:

  docker-compose logs flaresolverr
  

Other Resources

• MCP Protocol Documentation
• TypeScript Type Definitions

📄 License

This project is licensed under the ISC license.