SearxNG MCP Server logo

SearxNG MCP Server

by Sacode

SearxNG MCP Server provides web search capabilities using SearxNG for AI assistants like Claude. It offers a simple and efficient way for Large Language Models to search the web without tracking users.

View on GitHub

Last updated: N/A

What is SearxNG MCP Server?

This is a Model Context Protocol (MCP) server that connects to SearxNG, a privacy-respecting metasearch engine, allowing Large Language Models (LLMs) to search the web efficiently.

How to use SearxNG MCP Server?

The server can be integrated with MCP-compatible applications like Claude using configuration files (.clauderc) that specify the command and arguments to run the server. It can be run using pipx, uvx, Python with pip, or Docker. Configuration is done using environment variables.

Key features of SearxNG MCP Server

  • Privacy-focused web search through SearxNG

  • Simple API for LLM integration

  • Compatible with Claude Desktop and other MCP-compliant clients

  • Configurable search parameters

  • Clean, formatted search results optimized for LLMs

Use cases of SearxNG MCP Server

  • Enabling web search for AI assistants

  • Providing privacy-respecting search for LLMs

  • Integrating web search into MCP-compatible applications

  • Using SearxNG as a search backend for AI models

FAQ from SearxNG MCP Server

What is SearxNG?

SearxNG is a privacy-respecting metasearch engine that aggregates results from various search engines without tracking users.

What is MCP?

MCP stands for Model Context Protocol, a protocol for communication between Large Language Models and external tools or services.

How do I configure the SearxNG URL?

You can configure the SearxNG URL using the SEARXNG_MCP_SEARXNG_URL environment variable.

What transport protocols are supported?

The server supports STDIO (default) and SSE (Server-Sent Events) transport protocols.

How do I use the SSE transport protocol?

Set the TRANSPORT_PROTOCOL environment variable to sse and ensure your MCP client supports SSE. The server will be accessible via HTTP at http://localhost:8000 by default.