用于谷歌搜索的模型上下文协议服务器。一款基于 Playwright 的模型上下文协议 (MCP) 工具，可绕过搜索引擎反机器人机制，执行谷歌搜索并提取结果，为 Claude 和 Cursor 等人工智能助手提供实时搜索功能。
反机器人绕过：使用浏览器指纹欺骗和真实用户行为模拟来避免检测
自动验证码处理：遇到验证码时切换到头模式，允许用户完成验证
状态持久性：保存浏览器会话状态，以减少重复验证的需要
适应性：使用多种选择器组合来适应 Google 搜索页面的变化
MCP 集成：实现模型上下文协议，以便轻松与 AI 助手集成
多语言支持：支持不同语言和地区的搜索结果
Related MCP server: Playread
安装 - 全球
使用 npx 运行 MCP 服务器：
npx -y @mcp-server/google-search-mcp@latest在你的 Cursor IDE 中
前往Cursor Settings > MCP
点击+ Add New MCP Server
填写表格：
名称： google-search （或您喜欢的任何名称）
类型： command
命令： npx -y @mcp-server/google-search-mcp@latest
安装 - 特定于项目
将.cursor/mcp.json文件添加到您的项目：
{
"mcpServers": {
"google-search": {
"command": "npx",
"args": [
"-y",
"@mcp-server/google-search-mcp@latest"
]
}
}
}yarn installyarn buildyarn start要调试服务器，您可以使用MCP 检查器：
# First build the project
yarn build
# Start the MCP Inspector and server
npx @modelcontextprotocol/inspector node dist/index.js搜索工具接受以下参数：
query （必需）：搜索查询字符串
limit （可选）：返回的搜索结果数量，默认为 10
timeout （可选）：搜索操作超时（以毫秒为单位），默认值为 60000
language （可选）：搜索结果的语言，例如 zh-CN、en-US，默认为 zh-CN
region （可选）：搜索结果的地区，例如 cn、com、co.jp，默认为 cn
该工具使用 Playwright 控制 Chromium 浏览器执行 Google 搜索
它通过浏览器指纹欺骗和真实用户行为模拟来避免机器人检测
遇到CAPTCHA验证时，自动切换到头显模式供用户完成
它提取搜索结果并以结构化格式返回
它保存浏览器状态以便在后续搜索中重复使用
默认情况下，浏览器状态保存在用户主目录下的.google-search-browser-state.json文件中。您可以通过参数修改此路径。
您可以通过参数指定搜索结果的语言和地区：
// English (US) search results
"language": "en-US", "region": "com"
// Japanese search results
"language": "ja-JP", "region": "co.jp"
// Chinese (Simplified) search results
"language": "zh-CN", "region": "cn"首次使用时，如遇到CAPTCHA验证，系统将自动切换至头显模式，以便您完成验证
验证后系统会保存状态文件，使后续搜索更加顺畅
过于频繁的搜索请求可能会触发 Google 的速率限制机制
本工具仅供学习研究之用，请遵守Google的服务条款
麻省理工学院
本工具仅供学习研究之用。使用本工具访问谷歌或其他搜索引擎时，请遵守相关服务条款及相关法律法规。作者对因使用本工具而产生的任何问题概不负责。
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | 搜索查询字符串 | |
| limit | No | 返回的搜索结果数量，默认为10 | |
| timeout | No | 搜索操作的超时时间(毫秒)，默认为60000 | |
| language | No | 搜索结果的语言，例如 zh-CN, en-US 等，默认为 zh-CN | |
| region | No | 搜索结果的地区，例如 cn, com, co.jp 等，默认为 cn |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
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?
Tool has no description.
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?
Tool has no description.
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?
Tool has no description.
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?
Tool has no description.
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?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
searchScored across 1 tool
With only one tool, there is no possibility of ambiguity or overlap between tools. The single tool 'search' has a distinct and clear purpose by default.
A single tool inherently has perfect naming consistency, as there are no other tools to compare it against. The name 'search' follows a simple, clear verb pattern.
One tool is too few for a server named 'Google Search MCP', which suggests a domain that might benefit from more functionality (e.g., advanced search options, filtering, or result management). A single tool feels thin and under-scoped for such a purpose.
The server is severely incomplete for a search domain. With only a basic 'search' tool and no description, it lacks obvious features like pagination, filtering by date or type, or handling search results, which are essential for effective search operations.
Search Google straight from your AI agent. Web results, images, videos, news, products, scholarly ar
Web search, browser automation, scraping, crawling and CAPTCHA solving for AI agents.
Google search, news, maps, scholar and public webpages for AI agents, with Markdown results.
Real-time web search for AI agents: ranked results, source URLs, and optional AI answers.
