MCP.so
ログイン
S

SearXNG MCP server

@nik-sharky

基本情報

カテゴリ

検索

トランスポート

stdio

作者

nik-sharky

投稿者

Nik

SearXNG MCP server について

A Model Context Protocol (MCP) server that provides web search capabilities by integrating with a SearXNG instance.

#search#searxng#self-hosted-search

SearXNG MCP server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "searxng-mcp-nik-sharky": {
      "command": "python",
      "args": [
        "searxng_mcp.py",
        "--transport",
        "http",
        "--port",
        "32123",
        "--searxng",
        "http://searx.lan"
      ]
    }
  }
}

SearXNG MCP server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is SearXNG MCP server?

A Model Context Protocol (MCP) server that provides web search capabilities by integrating with a SearXNG instance. It aggregates results from over 130 engines and supports advanced search parameters for precise queries.

How to use SearXNG MCP server?

Install Python 3.10+, clone the repository, and install dependencies with pip install -r requirements.txt. Run the server using python searxng_mcp.py --transport http --port 32123 --searxng http://your-searxng-instance:8080. Configuration can be set via environment variables (SEARXNG_URL, MCP_PORT, MCP_HOST) or command-line arguments. Docker is also supported for containerized deployment.

Key features of SearXNG MCP server

  • Aggregated web search across 130+ engines.
  • Programmatic discovery of categories and engines.
  • Supports stdio, HTTP, and SSE transports.
  • Flexible configuration via environment variables and CLI.
  • Advanced parameters: categories, engines, language, pageno, time_range, safesearch.

Use cases of SearXNG MCP server

  • Integrate aggregated web search into AI agents or assistants.
  • Retrieve specialized content (images, news, science, maps) programmatically.
  • Dynamically adapt to available engines and categories via discovery tools.
  • Deploy a stateless search service for scalable, multi-client environments.

よくある質問

1

What prerequisites are needed to run SearXNG MCP server?

Python 3.10+ and a running SearXNG instance (accessible via HTTP) are required.

2

How do I configure the server?

Set `SEARXNG_URL`, `MCP_PORT`, and `MCP_HOST` in a `.env` file, or pass them as command-line arguments (e.g., `--searxng http://instance:8080`).

3

What transport options are available?

The server supports `stdio` (default for some clients), `http` (stateless HTTP), and `sse` (Server-Sent Events) transports.

4

How can I filter search results?

Use the `categories` (e.g., `news,it`), `engines` (e.g., `google,wikipedia`), `language`, `pageno`, `time_range` (`day`, `month`, `year`), and `safesearch` (0–2) parameters.

5

Can I discover available categories and engines dynamically?

Yes, the `searxng_get_info` tool returns the list of enabled categories and engines from your SearXNG instance.

コメント

「検索」の他のコンテンツ