G-Search MCP
G-搜索 MCP
强大的 Google 搜索 MCP 服务器,可同时使用多个关键字进行并行搜索。
该项目由google-search修改而来。
优势
并行搜索:支持在Google上同时使用多个关键字进行搜索,提高搜索效率
浏览器优化:在单个浏览器实例中打开多个选项卡,以实现高效的并行搜索
自动验证处理:智能检测验证码,并在需要时启用可见浏览器模式进行用户验证
用户行为模拟:模拟真实用户浏览模式,降低被搜索引擎检测到的可能性
结构化数据:以 JSON 格式返回结构化搜索结果,以便于处理和分析
可配置参数:支持各种参数配置,如搜索结果限制、超时设置、区域设置等。
Related MCP server: Enhanced Google Search MCP
快速入门
直接用npx运行:
npx -y g-search-mcp首次设置 - 通过在终端中运行以下命令来安装所需的浏览器:
npx playwright install chromium调试模式
使用--debug选项以调试模式运行(显示浏览器窗口):
npx -y g-search-mcp --debug配置 MCP
在 Claude Desktop 中配置此 MCP 服务器:
MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%/Claude/claude_desktop_config.json
{
"mcpServers": {
"g-search": {
"command": "npx",
"args": ["-y", "g-search-mcp"]
}
}
}特征
search- 使用多个关键字执行 Google 搜索并返回结果使用 Playwright 浏览器进行搜索
支持以下参数:
queries:要执行的搜索查询数组(必需参数)limit:每个查询返回的最大结果数,默认值为 10timeout:页面加载超时时间(毫秒),默认为 60000(60 秒)noSaveState:是否避免保存浏览器状态,默认为 falselocale:搜索结果的语言环境设置,默认为 en-USdebug:是否启用调试模式(显示浏览器窗口),覆盖命令行中的 --debug 标志
使用示例:
Use the search tool to search for "machine learning" and "artificial intelligence" on Google响应示例:
{
"searches": [
{
"query": "machine learning",
"results": [
{
"title": "What is Machine Learning? | IBM",
"link": "https://www.ibm.com/topics/machine-learning",
"snippet": "Machine learning is a branch of artificial intelligence (AI) and computer science which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy."
},
...
]
},
{
"query": "artificial intelligence",
"results": [
{
"title": "What is Artificial Intelligence (AI)? | IBM",
"link": "https://www.ibm.com/topics/artificial-intelligence",
"snippet": "Artificial intelligence leverages computers and machines to mimic the problem-solving and decision-making capabilities of the human mind."
},
...
]
}
]
}使用技巧
处理特殊网站场景
调整搜索参数
搜索结果数量: 更多搜索结果:
Please return the top 20 search results for each keyword这将设置
limit: 20参数。增加超时时间:对于加载缓慢的情况:
Please set the page loading timeout to 120 seconds这会将
timeout参数调整为 120000 毫秒。
区域设置调整
更改搜索区域:指定不同的区域设置:
Please use Chinese locale (zh-CN) for searching这将设置
locale: "zh-CN"参数。
调试和故障排除
启用调试模式
动态调试激活:在特定搜索操作期间显示浏览器窗口:
Please enable debug mode for this search operation即使服务器启动时没有使用
--debug标志,也会设置debug: true。
安装
先决条件
Node.js 18 或更高版本
NPM 或 Yarn
从源安装
克隆存储库:
git clone https://github.com/jae-jae/g-search-mcp.git
cd g-search-mcp安装依赖项:
npm install安装 Playwright 浏览器:
npm run install-browser构建服务器:
npm run build发展
自动重建(开发模式)
npm run watch使用 MCP Inspector 进行调试
npm run inspector相关项目
fetcher-mcp :一款功能强大的 MCP 服务器,可使用 Playwright 无头浏览器抓取网页内容。它具备智能内容提取、并行处理、资源优化等功能,是理想的网页内容抓取工具。
执照
根据MIT 许可证授权
Available Tools
1 toolsearchC
Search on Google for multiple keywords and return the results
| Name | Required | Description | Default |
|---|---|---|---|
| debug | No | Whether to enable debug mode (showing browser window), overrides the --debug command line flag if specified | |
| limit | No | Maximum number of results to return per query (default: 10) | |
| locale | No | Locale setting for search results (default: en-US) | |
| noSaveState | No | Whether to avoid saving browser state (default: false) | |
| queries | Yes | Array of search queries to perform | |
| timeout | No | Page loading timeout in milliseconds (default: 60000) |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden for behavioral disclosure. It mentions the action (search Google) and output (return results) but omits critical behavioral details like rate limits, authentication requirements, error handling, or what format results are returned in. This leaves significant gaps for an AI agent to understand tool behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise - a single sentence that efficiently communicates the core functionality. Every word earns its place with no redundant information. The structure is front-loaded with the essential action and purpose.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a tool with 6 parameters, no annotations, and no output schema, the description is insufficiently complete. It doesn't explain what format results are returned in, how errors are handled, or provide context about the search operation's limitations. The agent would need to guess about important behavioral aspects.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, providing comprehensive parameter documentation. The description adds minimal value beyond the schema, only implying that 'queries' parameter contains search keywords. No additional semantic context is provided about parameter interactions or usage patterns.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Search on Google for multiple keywords and return the results.' It specifies the verb (search), resource (Google), and scope (multiple keywords). However, with no sibling tools mentioned, there's no opportunity to differentiate from alternatives, preventing a perfect score.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It doesn't mention prerequisites, constraints, or typical use cases. While there are no sibling tools to contrast with, the description lacks any contextual usage information.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
1 tool update
v1.0.0- First observed
search
TDQS
Scored across 1 tool
With only one tool, there is no possibility of confusion or overlap between tools, making disambiguation perfect. The single tool has a clear and distinct purpose without any ambiguity.
With only one tool, naming consistency is inherently perfect as there are no other tools to compare against. The tool name 'search' follows a simple verb pattern, which is appropriate for its function.
A single tool is too few for a search server, as it lacks basic operations like refining searches, handling different result types, or managing search history. This minimal set feels thin and incomplete for the apparent scope of a search service.
The tool set is severely incomplete for a search domain, offering only a basic search function without any support for advanced features like filtering, pagination, or accessing specific result details. This will likely cause agent failures when more complex search tasks are required.
Maintenance
Related MCP Connectors
MCP server for Google search results via SERP API
MCP server for 500+ pay-per-call web scraping, search, social, business, and financial data tools.
Official SerpApi MCP server for Google, Bing, and other search engines.
All HasData scraping tools in one MCP server: Google, TikTok, Instagram, maps, e-commerce and more.
Related MCP Servers
- AlicenseAqualityCmaintenanceAn MCP server that provides Google Search functionality with automatic API key rotation and intelligent quota management, enabling natural language search queries with advanced filtering options.111 npm1MIT
- AlicenseAqualityDmaintenanceA powerful Model Context Protocol (MCP) server that provides enhanced Google search capabilities with advanced anonymization and anti-detection features.14MIT
- AlicenseNot gradedqualityDmaintenanceAn MCP server that provides web search and content fetching using real Chrome browsers via CDP, bypassing bot detection and requiring no API keys.20MIT
- FlicenseAqualityDmaintenanceMCP server that provides automated Google search capabilities, allowing query execution, result limiting, and optional related keywords retrieval.1-
Appeared in Searches
- A powerful MCP server for web search and information gathering
- Techniques for Scraping Dynamic Websites with JavaScript and Handling CAPTCHA or Proxy Issues
- Information about Meta ads or advertising on Meta platforms
- A list of all Minecraft plugins that can use SSE
- Information about DeepSearch or related platforms/tools