SerpApi MCP Server

Name: SerpApi MCP Server
Author: ilyazub

@ilyazub

リポジトリホームページ

SerpApi MCP Server について

SerpApi MCP Server for Google and other search engine results

#serpapi #mcp-server #api-integration

基本情報

カテゴリ

検索

ライセンス

MIT

ランタイム

python

トランスポート

stdio

公開者

ilyazub

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": {
    "serpapi-mcp-server": {
      "command": "python",
      "args": [
        "server.py"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

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 SerpApi MCP Server?

A Model Context Protocol (MCP) server that integrates with SerpApi to provide comprehensive search engine results and data extraction from multiple search engines.

How to use SerpApi MCP Server?

Use the hosted service at mcp.serpapi.com with your API key in the URL path or Authorization header. Configure your MCP client (e.g., Claude Desktop) with a JSON entry pointing to https://mcp.serpapi.com/YOUR_SERPAPI_API_KEY/mcp. For self-hosting, clone the repository, run uv sync && uv run src/server.py, and set http://localhost:8000/YOUR_SERPAPI_API_KEY/mcp in your client.

Key features of SerpApi MCP Server

Multi-engine search (Google, Bing, Yahoo, DuckDuckGo, YouTube, eBay, and more)
Engine-specific parameter schemas exposed as MCP resources
Real-time weather and stock market data via search queries
Dynamic detection and formatting of result types (answer boxes, organic, news, images, shopping)
Flexible compact or complete JSON response modes
Optional interactive UI tools (search_table and search_dashboard) for supporting hosts

Use cases of SerpApi MCP Server

Performing web searches through an AI assistant that uses MCP
Extracting structured search data (e.g., local business info, news results) for analysis
Fetching real-time weather or stock information without a separate API call
Building conversational agents that present search results as interactive tables or dashboards

FAQ from SerpApi MCP Server

How does authentication work?

Two methods: path-based (include API key in URL: /YOUR_API_KEY/mcp) or header-based (Authorization: Bearer YOUR_API_KEY).

What languages or runtimes are required?

Python 3.13+ is required for self-hosting. The hosted service requires only an API key and an MCP client.

Where does the data come from?

All search data is fetched from SerpApi, which aggregates results from supported search engines based on your query and parameters.

What are the known limits or error cases?

Common issues include a missing or invalid API key, rate limit exceeded (wait or upgrade plan), or "no results" for certain queries or engines.

Which transports are supported?

The server uses HTTP transport with a streamable HTTP endpoint (e.g., https://mcp.serpapi.com/YOUR_API_KEY/mcp). Self-hosted uses the same configuration on localhost:8000.