Local Web Search

by fradser

6 stars
390 downloads
Not rated
GitHub

About

Integrates with a headless browser to perform Google searches and extract web content, enabling local web search capabilities for up-to-date information retrieval and research tasks.

Details

Author
fradser
Repository
FradSer/mcp-server-local-web-search
GitHub stars
6
Downloads
390
License
MIT License
Categories
Productivity, Developer Tools, Design, AI, Search, Infrastructure, Frontend, Knowledge Base
Tags
#web

- Perform web searches with customizable result limits
- Extract and process content from web pages
- Return structured results with titles, URLs, and descriptions
- Support for content truncation and domain filtering
- Clean content extraction using Readability
- Headless browser operation for improved performance

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

  1. Download and install Highlight from highlightai.com/download
  2. Navigate to the plugins tab and select "Add Custom Plugin"
  3. Configure the plugin with the settings below
    Plugin Name Local Web Search
    Command (node, npx, python, etc.) npx
    Arguments
    • Argument 1 -y
    • Argument 2 @highlight/mcp-server

    Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.

  4. Enable "Start Automatically" if you want the plugin to start when Highlight launches

From the repository

Run the setup script to configure the MCP server:

bun run setup.ts

This will add the server to your Claude MCP configuration.

local_web_search

Performs web search and returns results with title, URL and description. Parameters: query (required), excludeDomains (optional list), limit (optional number), truncate (optional number), show (optional boolean), proxy (optional string)

1. local_web_search
- Performs web search and returns results with title, URL and description
- Parameters:
- query: Search query to find relevant content (required)
- excludeDomains: List of domains to exclude from search results (default: [])
- limit: Maximum number of results to return (default: 5)
- truncate: Maximum length of content to return per result (default: 4000)
- show: Show browser window for debugging (default: false)
- proxy: Proxy server to use for requests (optional)

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "local web search": {
            "env": {},
            "args": [
                "-y",
                "@highlight/mcp-server"
            ],
            "command": "npx"
        }
    }
}

Linux

{
    "env": [],
    "args": [
        "-y",
        "@highlight/mcp-server"
    ],
    "command": "npx"
}

Macos

{
    "env": [],
    "args": [
        "-y",
        "@highlight/mcp-server"
    ],
    "command": "npx"
}

Windows

{
    "env": [],
    "args": [
        "/c",
        "npx",
        "-y",
        "@highlight/mcp-server"
    ],
    "command": "cmd"
}

mcp-server-local-web-search

Twitter Follow

An MCP server for performing local web searches. This server provides tools to search and extract content from web pages through the Model Context Protocol.

Features

- Perform web searches with customizable result limits
- Extract and process content from web pages
- Return structured results with titles, URLs, and descriptions
- Support for content truncation and domain filtering
- Clean content extraction using Readability
- Headless browser operation for improved performance

Installation

To install dependencies:

bun install

Setup

Run the setup script to configure the MCP server:

bun run setup.ts

This will add the server to your Claude MCP configuration.

Available Tools

1. local_web_search
- Performs web search and returns results with title, URL and description
- Parameters:
- query: Search query to find relevant content (required)
- excludeDomains: List of domains to exclude from search results (default: [])
- limit: Maximum number of results to return (default: 5)
- truncate: Maximum length of content to return per result (default: 4000)
- show: Show browser window for debugging (default: false)
- proxy: Proxy server to use for requests (optional)

Requirements

- Bun runtime
- Node.js TypeScript support

Development

This project uses:

- Bun as the JavaScript runtime
- TypeScript for type safety
- Model Context Protocol SDK for server implementation
- @egoist/local-web-search for web search (using playwright-core)
- Readability for content extraction

Contributors

- egoist - Original local web search author - FradSer - Original author - TheSethRose - Playwright integration and performance improvements

License

MIT License

This project was created using bun init in bun v1.2.2. Bun is a fast all-in-one JavaScript runtime.

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.