searxng-simple-mcp
The SearxNG MCP Server enables AI assistants to perform web searches through the SearxNG engine, maintaining user privacy and optimizing for large language models. It is designed for easy integration, offering flexibility through various configuration and installation options.
Overview
This project implements an MCP server that connects to SearxNG, a privacy-respecting metasearch engine. It provides seamless web search capabilities for Large Language Models (LLMs) like Claude, without user tracking. Key features include a privacy-focused web search, a simple API for LLM integration, compatibility with MCP-compliant clients, configurable search parameters, and clean search results optimized for LLMs. The server can be configured using environment variables, and it supports multiple installation methods, including pip, Docker, and Docker Compose. The available transport protocols are STDIO and SSE, suitable for different levels of integration and real-time updates.