A
licenseNot graded
qualityD
maintenanceEnables AI models to search the web for current information before generating responses, with features for conditional searching, geographic customization, and automatic citations.
1
MIT
mcp-grok-search
by Cedriccmh
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcp-grok-searchsearch for recent advances in renewable energy"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
mcp-grok-search
English
An MCP (Model Context Protocol) server that provides real-time web search capabilities via the Grok API. Enables AI assistants like Claude to search the internet through Grok's built-in web browsing.
Features
Real-time web search powered by Grok AI
Uses the Responses API, which is the recommended path for Grok web search
grok_web_searchsupports an optionaleffortspreset:low,medium, orhighefforts=lowusesgrok-4.20-reasoningefforts=medium/highusegrok-4.20-multi-agentwith official reasoning presetsConfigurable API endpoint, model, and timeout
Stdio transport for seamless MCP integration
Quick Start
Prerequisites
Node.js >= 18
A Grok API key (get one at x.ai)
Install & Build
git clone https://github.com/Cedriccmh/mcp-grok-search.git
cd mcp-grok-search
npm install
npm run buildEnvironment Variables
Variable | Description | Default |
| (Required) Your Grok API key | — |
| API endpoint URL |
|
| Default model used when |
|
| Request timeout in ms |
|
Run
GROK_API_KEY=your-key-here npm startClaude Desktop Configuration
Add to your claude_desktop_config.json:
{
"mcpServers": {
"grok-web-search": {
"command": "node",
"args": ["path/to/mcp-grok-search/dist/index.js"],
"env": {
"GROK_API_KEY": "your-key-here"
}
}
}
}Tech Stack
TypeScript
@modelcontextprotocol/sdk — MCP server framework
Zod — Input schema validation
License
MIT
Related MCP server: Gemini Search MCP
中文
一个基于 Grok API 的 MCP（Model Context Protocol）服务器，提供实时网络搜索能力。让 Claude 等 AI 助手能够通过 Grok 内置的网页浏览功能搜索互联网。
特性
基于 Grok AI 的实时网络搜索
使用 Responses API，符合 Grok 当前推荐的搜索接入方式
grok_web_search工具支持可选的efforts预设：low、medium、highefforts=low使用grok-4.20-reasoningefforts=medium/high使用grok-4.20-multi-agent和官方推荐的reasoning.effort可配置 API 端点、模型和超时时间
Stdio 传输，无缝对接 MCP 协议
快速开始
前置要求
Node.js >= 18
Grok API 密钥（在 x.ai 获取）
安装与构建
git clone https://github.com/Cedriccmh/mcp-grok-search.git
cd mcp-grok-search
npm install
npm run build环境变量
变量 | 说明 | 默认值 |
| （必填） Grok API 密钥 | — |
| API 端点 URL |
|
| 不传 |
|
| 请求超时时间（毫秒） |
|
运行
GROK_API_KEY=your-key-here npm startClaude Desktop 配置
将以下内容添加到 claude_desktop_config.json：
{
"mcpServers": {
"grok-web-search": {
"command": "node",
"args": ["path/to/mcp-grok-search/dist/index.js"],
"env": {
"GROK_API_KEY": "your-key-here"
}
}
}
}技术栈
TypeScript
@modelcontextprotocol/sdk — MCP 服务端框架
Zod — 输入参数校验
许可证
MIT
Available Tools
1 toolgrok_web_searchGrok Web SearchARead-only
Search the web using Grok AI with real-time internet access. Grok autonomously searches, browses pages, and synthesizes results. Returns comprehensive answers. Use for: any scenario requiring real-time web information.
ParametersJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | The search query in natural language | |
| efforts | No | Optional search scale preset. low uses grok-4.20-reasoning, medium/high use grok-4.20-multi-agent with official reasoning effort presets. |
TDQS
A4.3/5.0
Behavior4/5
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already indicate read-only and open-world behavior. The description adds that Grok 'autonomously searches, browses pages, and synthesizes results,' giving insight into the autonomous nature and real-time access beyond the structured hints.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Conciseness5/5
Is the description appropriately sized, front-loaded, and free of redundancy?
Three tight sentences pack the action, behavior, output, and use case without redundant phrasing. The description is front-loaded with the core purpose and remains concise.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Completeness5/5
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the annotations, full schema coverage, and no output schema, the description covers the tool's purpose, behavior, and usage adequately. The absence of an output schema is compensated by 'Returns comprehensive answers,' which sets expectations.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Parameters3/5
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%; both parameters have descriptive comments (query and efforts). The description itself does not elaborate on parameter details, but the schema handles this, so the baseline of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Purpose5/5
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's function: 'Search the web using Grok AI with real-time internet access.' It further specifies autonomous browsing and synthesis, making the purpose specific and distinct even without sibling tools.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Usage Guidelines4/5
Does the description explain when to use this tool, when not to, or what alternatives exist?
It explicitly provides a usage guideline: 'Use for: any scenario requiring real-time web information.' While it doesn't mention when-not to use, 'any scenario' is broad and covers the main use case. Since no sibling tools exist, the lack of alternatives is acceptable.
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
grok_web_search
TDQS
A4.4/5.0
Scored across 1 tool
Disambiguation5/5
Only one tool exists, so there is no possibility of confusion between tools. The tool's purpose is clear and distinct, making selection unambiguous.
Naming Consistency5/5
The single tool name follows a clear verb_noun pattern (grok_web_search), which is consistent and predictable within the set.
Tool Count3/5
With only one tool, the set is at the lower boundary of what is typical. While a focused search server could reasonably have just one search tool, the minimal count feels thin and may limit flexibility.
Completeness5/5
The tool fully covers the search capability for this server's stated purpose, providing real-time web search with autonomous browsing and synthesis. No obvious missing operations are evident.
Maintenance
ActivityInactive
ResponsivenessNo issues
Related MCP Connectors
Enable AI assistants to perform web searches using Perplexity's Sonar Pro.
Provides AI assistants with access to Seltz's powerful Web Search capabilities.
The best web search for your AI Agent
Web research for agents: quality-scored Google search, webpage extraction, and deep research.
Related MCP Servers
- AlicenseBqualityDmaintenanceEnables AI-powered web searches using Google's Gemini 2.5 models with Google Search Grounding. Supports multiple Gemini models (Flash, Flash-Lite, Pro) with configurable thinking budget and web search capabilities.112 npm2MIT
- AlicenseNot gradedqualityDmaintenanceEnables web search via DuckDuckGo and web content fetching through the Model Context Protocol, allowing AI assistants to search and retrieve web page content.13 npmMIT
- AlicenseNot gradedqualityDmaintenanceEnables real-time web and Twitter/X search via Grok, returning structured results with source URLs, confidence scores, and key points. Supports multiple output modes, language options, and time range filtering.22 npm4MIT