{"id":"mcp-searxng","name":"MCP SearXNG","homepage":"https://github.com/ihor-sokoliuk/mcp-searxng","repo_url":"https://github.com/ihor-sokoliuk/mcp-searxng","category":"search","subcategories":["web-search","content-extraction","privacy"],"tags":["searxng","web-search","privacy","metasearch","url-reader","markdown","self-hosted"],"what_it_does":"An MCP server that connects AI agents to a SearXNG metasearch engine instance for privacy-respecting web search and URL content extraction, returning results and page content in clean markdown format.","use_cases":["Give AI agents real-time web search capabilities without sending queries to Google or Bing directly","Extract and convert web page content to markdown for AI summarization or analysis","Build research agents that need paginated, time-filtered, or language-specific web results","Self-host a search backend for an AI stack with full query privacy"],"not_for":["Teams without a running SearXNG instance — requires self-hosting the search engine","Scenarios needing guaranteed result freshness from a specific search index (metasearch quality varies by SearXNG config)","High-volume crawling — SearXNG instances have their own rate limits from upstream engines"],"best_when":"You already run SearXNG (or want to) and need a privacy-preserving, self-hosted search tool for your AI agents.","avoid_when":"You need a managed, SLA-backed search API without self-hosting overhead — use Brave Search MCP or Tavily instead.","alternatives":["mcp-brave-search","mcp-tavily","mcp-exa"],"af_score":73.2,"security_score":72.0,"reliability_score":null,"package_type":"mcp_server","discovery_source":["github","github_awesome","npm"],"priority":"low","status":"evaluated","version_evaluated":"latest","last_evaluated":"2026-03-01T09:50:05.891122+00:00","performance":{"latency_p50_ms":null,"latency_p99_ms":null,"uptime_sla_percent":null,"rate_limits":null,"data_source":"llm_estimated","measured_on":null}}