# Grok Chat MCP server

> Enables AI clients to interact with the xAI Grok API for chat completions, vision analysis, web and X search, and text embeddings.

- Repository: https://github.com/codeChap/mcp-server-grok-chat
- Page: https://hostedmcp.org/servers/codechap/mcp-server-grok-chat
- Transport: —
- Stars: 0
- Latest version: 0.2.0

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `RUST_LOG` | no | no |  |

## Tools (17)

- `chat`
- `chat_with_vision`
- `chat_with_search`
- `embedding`
- `list_models`
- `prompt`
- `model`
- `system_prompt`
- `messages`
- `temperature`
- `max_tokens`
- `response_schema`
- `reasoning_effort`
- `image_url`
- `detail`
- `search_type`
- `input`

## FAQ

### What capabilities does the Grok MCP server provide?

The server provides tools for chat completions, vision analysis, live web and X (Twitter) search, text embeddings, and model listing.

### How can I control the number of agents used in multi-agent models?

You can use the reasoning_effort parameter; for multi-agent models, low/medium settings use 4 agents, while high/xhigh settings use 16 agents.

### What is required to set up this server?

You need Rust (edition 2024) and an xAI API key from the xAI console.
