xAI MCP Server
Integrates xAI's Grok APIs into Claude Code to enable image and video generation, real-time web searches, and multi-modal image analysis. It provides a suite of tools for interacting with Grok models directly through natural language prompts during a Claude session.
README
xAI MCP Server
A Model Context Protocol (MCP) server that brings xAI's Grok APIs to Claude Code. Generate images, chat with Grok, analyze images, search the web, and create videos—all from natural language prompts in your Claude Code session.
Features
| Tool | Description |
|---|---|
generate_image |
Generate images using Grok Imagine |
chat |
Chat with Grok models (grok-3, grok-4, grok-3-mini) |
analyze_image |
Analyze and describe images with Grok Vision |
live_search |
Real-time web, news, and X/Twitter search |
generate_video |
Generate videos from text prompts |
Prerequisites
- Node.js 18.0.0 or higher
- xAI API Key from x.ai/api
- Claude Code installed
Installation
Option 1: Quick Install (Recommended)
curl -fsSL https://raw.githubusercontent.com/joemccann/xai-mcp-server/main/install.sh | bash
This installs the server and automatically configures Claude Code. You'll be prompted for your xAI API key.
Option 2: npx from GitHub
No installation needed - runs directly from GitHub:
npx github:joemccann/xai-mcp-server
Option 3: npm Global Install
npm install -g @joemccann/xai-mcp-server
Option 4: Clone and Build
git clone https://github.com/joemccann/xai-mcp-server.git
cd xai-mcp-server
npm install
Configuration for Claude Code
Step 1: Get Your xAI API Key
- Go to x.ai/api
- Sign up or log in
- Create an API key
- Copy the key (starts with
xai-)
Step 2: Configure Claude Code
Note: If you used the Quick Install (Option 1), this is already done for you.
Add the MCP server to your Claude Code settings file:
Location: ~/.claude/settings.json
Using npx from npm (no install needed):
{
"mcpServers": {
"xai": {
"command": "npx",
"args": ["-y", "@joemccann/xai-mcp-server"],
"env": {
"XAI_API_KEY": "xai-your-api-key-here"
}
}
}
}
Using npx from GitHub:
{
"mcpServers": {
"xai": {
"command": "npx",
"args": ["-y", "github:joemccann/xai-mcp-server"],
"env": {
"XAI_API_KEY": "xai-your-api-key-here"
}
}
}
}
If installed globally via npm:
{
"mcpServers": {
"xai": {
"command": "xai-mcp-server",
"env": {
"XAI_API_KEY": "xai-your-api-key-here"
}
}
}
}
If cloned/installed locally:
{
"mcpServers": {
"xai": {
"command": "node",
"args": ["~/.xai-mcp-server/dist/index.js"],
"env": {
"XAI_API_KEY": "xai-your-api-key-here"
}
}
}
}
Step 3: Restart Claude Code
Restart Claude Code to load the new MCP server. You should see the xAI tools available.
Usage
Once configured, you can use natural language to invoke xAI capabilities:
Image Generation
Generate an image of a cyberpunk cityscape at night with neon lights.
Using grok imagine, create a watercolor painting of a mountain landscape.
Generate 3 variations of a logo for a coffee shop called "Bean There".
Chat with Grok
Ask Grok to explain the theory of relativity in simple terms.
Have Grok write a haiku about programming.
Image Analysis
Analyze this image and describe what you see: https://example.com/photo.jpg
What text is visible in this screenshot: [image URL]
Live Search
Search for the latest news about SpaceX launches.
Find recent tweets about the new iPhone release.
Search the web for Python best practices 2024.
Video Generation
Generate a 5-second video of clouds moving across a blue sky.
Create a video animation of a bouncing ball.
Tool Reference
generate_image
Generate images from text descriptions using Grok Imagine.
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
prompt |
string | Yes | - | Text description of the image |
n |
number | No | 1 | Number of images (1-10) |
model |
string | No | grok-2-image |
Image generation model |
aspect_ratio |
string | No | - | Aspect ratio (e.g., "16:9", "1:1", "4:3") |
response_format |
string | No | url |
Output format: "url" or "b64_json" |
Example Response:
{
"success": true,
"images": [
{
"index": 1,
"url": "https://api.x.ai/images/generated/abc123.png",
"revised_prompt": "A detailed cyberpunk cityscape..."
}
]
}
chat
Chat with Grok language models.
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
message |
string | Yes | - | Message to send to Grok |
model |
string | No | grok-3 |
Model: grok-3, grok-4, grok-3-mini |
system_prompt |
string | No | - | System context/instructions |
temperature |
number | No | 0.7 | Sampling temperature (0-2) |
max_tokens |
number | No | - | Maximum response tokens |
analyze_image
Analyze images using Grok's vision capabilities.
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
image_url |
string | Yes | - | Image URL or base64 data URL |
prompt |
string | No | "Describe this image" | Question or instruction |
detail |
string | No | auto |
Detail level: "low", "high", "auto" |
model |
string | No | grok-2-vision-1212 |
Vision model |
live_search
Perform real-time web searches using Grok.
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
query |
string | Yes | - | Search query |
sources |
array | No | ["web"] |
Sources: "web", "news", "x" |
date_range |
object | No | - | Date filter: { start, end } (YYYY-MM-DD) |
max_results |
number | No | 10 | Maximum results (1-20) |
generate_video
Generate videos from text descriptions.
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
prompt |
string | Yes | - | Video description |
model |
string | No | grok-2-video |
Video generation model |
duration |
number | No | 5 | Duration in seconds (1-15) |
image |
string | No | - | Input image URL to animate |
video |
string | No | - | Input video URL to edit |
aspect_ratio |
string | No | - | Aspect ratio (e.g., "16:9") |
wait_for_completion |
boolean | No | true | Wait for video to finish |
Development
# Install dependencies
npm install
# Build TypeScript
npm run build
# Watch mode (rebuild on changes)
npm run dev
# Run the server directly
npm start
Project Structure
xai-mcp-server/
├── src/
│ ├── index.ts # MCP server entry point
│ ├── xai-client.ts # xAI API client with types
│ └── tools/
│ ├── generate-image.ts
│ ├── chat.ts
│ ├── vision.ts
│ ├── live-search.ts
│ └── generate-video.ts
├── dist/ # Compiled JavaScript
├── package.json
├── tsconfig.json
└── README.md
Troubleshooting
"XAI_API_KEY environment variable is required"
Ensure your ~/.claude/settings.json includes the env block with your API key:
"env": {
"XAI_API_KEY": "xai-your-key-here"
}
Tools not appearing in Claude Code
- Verify the path in
argsis absolute and correct - Ensure the project is built (
npm run build) - Restart Claude Code completely
- Check Claude Code logs for MCP connection errors
API errors
- Verify your API key is valid at x.ai
- Check you have sufficient API credits
- Some features may require specific API tier access
API Reference
This server uses the xAI API. For full API documentation, see:
License
MIT
Contributing
Contributions welcome! Please open an issue or submit a pull request.
Acknowledgments
- xAI for the Grok API
- Model Context Protocol for the MCP specification
- Anthropic for Claude Code
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.