SearXNG MCP Server

SearchTypeScript

Web search capabilities through SearXNG API integration

Available Tools

searxng_web_search

Execute web searches with pagination and filtering options

query: The search query string to be passed to external search servicespageno: (Optional) Search page number, starts at 1 (default: 1)time_range: (Optional) Filter results by time range - one of: 'day', 'month', 'year' (default: none)language: (Optional) Language code for results (e.g., 'en', 'fr', 'de') or 'all' (default: 'all')safesearch: (Optional) Safe search filter level (0: None, 1: Moderate, 2: Strict) (default: instance setting)

web_url_read

Read and convert the content from a URL to markdown

url: The URL to fetch and process

SearXNG MCP Server provides web search functionality by integrating with the SearXNG API, a privacy-respecting metasearch engine. It allows users to perform web searches with advanced filtering options including pagination, time range filtering, language selection, and safe search controls. This MCP server enables AI assistants to search the web for current information, helping them provide up-to-date responses on recent events, trending topics, or any information that might not be in their training data.

Overview

SearXNG MCP Server integrates with SearXNG, a privacy-focused metasearch engine, to provide web search capabilities to AI assistants. This integration allows AI models to search the web for current information, helping them provide more accurate and up-to-date responses.

Installation Options

There are several ways to install and use the SearXNG MCP Server:

Using NPX (Recommended)

Add the following configuration to your AI assistant's configuration file:

{
  "mcpServers": {
    "searxng": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-searxng"
      ],
      "env": {
        "SEARXNG_URL": "YOUR_SEARXNG_INSTANCE_URL"
      }
    }
  }
}

Using NPM Global Installation

First, install the package globally:

npm install -g mcp-searxng

Then add this to your configuration file:

{
  "mcpServers": {
    "searxng": {
      "command": "mcp-searxng",
      "env": {
        "SEARXNG_URL": "YOUR_SEARXNG_INSTANCE_URL"
      }
    }
  }
}

Using Docker

You can use the pre-built Docker image:

{
  "mcpServers": {
    "searxng": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "SEARXNG_URL",
        "isokoliuk/mcp-searxng:latest"
      ],
      "env": {
        "SEARXNG_URL": "YOUR_SEARXNG_INSTANCE_URL"
      }
    }
  }
}

Using Smithery (for Claude Desktop)

For Claude Desktop users, you can install via Smithery:

npx -y @smithery/cli install @ihor-sokoliuk/server-searxng --client claude

Configuration

SearXNG Instance URL

You need to set the SEARXNG_URL environment variable to point to a SearXNG instance. You can:

Use a public instance from searx.space
Run your own local instance (default URL is http://localhost:8080)

The server will use this instance to perform web searches.

Usage

Once configured, your AI assistant can use the following tools:

Web Search: Search the web with various filtering options
URL Content Reading: Fetch and convert web page content to markdown

You can ask your AI assistant to search for specific information, news, or articles, and it will use the SearXNG integration to provide current results.