Serper Search MCP Server

Name: Serper Search MCP Server
Author: NightTrek

@NightTrek

Repository

About Serper Search MCP Server

Un-official Serper Google search server for Cline and other MCP clients

Basic information

Config

No standard config provided

This server doesn't expose a parseable MCP config block in its README. See the repository for install instructions.

Repository

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Serper Search MCP Server?

A Model Context Protocol server that provides Google search capabilities through the Serper API, along with an AI-powered Deep Research tool. It enables integration of search and research functionality into MCP-enabled applications.

How to use Serper Search MCP Server?

Clone the repository, install dependencies with pnpm, build the server, then configure by adding your Serper API key in a .env file. Integrate with Claude Desktop by adding the server config to the claude_desktop_config.json file, specifying the path to the built index.js and the API key.

Key features of Serper Search MCP Server

Google search integration through Serper API
Rich search response: Knowledge Graph, organic results, People Also Ask, related searches
AI-powered Deep Research tool with multi-step iterative research
Configurable search parameters: country, language, result count, autocorrect
Secure API key handling
TypeScript support with full type definitions

Use cases of Serper Search MCP Server

Quickly retrieve organic search results and knowledge panels for any query
Perform deep, multi-step research on complex topics with citations
Integrate search functionality into MCP-enabled applications like Claude Desktop
Customize search by country, language, and number of results

FAQ from Serper Search MCP Server

How do I get a Serper API key?

What parameters does the Search tool accept?

The Search tool accepts query, numResults (default 10, max 100), gl (country code), hl (language code), autocorrect (default true), and type (currently only "search").

What are the depth levels for Deep Research?

Basic (3-5 sources, ~5 min), Standard (5-10 sources, ~10 min), Deep (10+ sources, ~15-20 min).

Does the server collect quality metrics?

Yes, the Deep Research tool includes integrated quality metrics for research process, performance, and usage patterns. See TELEMETRY.md for details. Metrics collection can be configured but is enabled by default.

What transport does the server use?

The README shows it is configured as an MCP server with the command pointing to the built index.js file; it uses stdio transport (as typical for MCP servers in Claude Desktop).