Grok MCP server
{
"mcpServers": {
"acedatacloud-grok": {
"command": "uvx",
"args": [
"mcp-grok"
],
"env": {
"ACEDATACLOUD_API_TOKEN": "<acedatacloud_api_token>",
"ACEDATACLOUD_API_BASE_URL": "https://..."
}
}
}
}{
"mcpServers": {
"acedatacloud-grok": {
"command": "uvx",
"args": [
"mcp-grok"
],
"env": {
"ACEDATACLOUD_API_TOKEN": "<acedatacloud_api_token>",
"ACEDATACLOUD_API_BASE_URL": "https://..."
}
}
}
}{
"mcpServers": {
"acedatacloud-grok": {
"command": "uvx",
"args": [
"mcp-grok"
],
"env": {
"ACEDATACLOUD_API_TOKEN": "<acedatacloud_api_token>",
"ACEDATACLOUD_API_BASE_URL": "https://..."
}
}
}
}{
"servers": {
"acedatacloud-grok": {
"command": "uvx",
"args": [
"mcp-grok"
],
"env": {
"ACEDATACLOUD_API_TOKEN": "${input:acedatacloud_api_token}",
"ACEDATACLOUD_API_BASE_URL": "${input:acedatacloud_api_base_url}"
}
}
},
"inputs": [
{
"id": "acedatacloud_api_token",
"type": "promptString",
"description": "Acedatacloud api token",
"password": true
},
{
"id": "acedatacloud_api_base_url",
"type": "promptString",
"description": "Acedatacloud api base url",
"password": true
}
]
}{
"mcpServers": {
"acedatacloud-grok": {
"command": "uvx",
"args": [
"mcp-grok"
],
"env": {
"ACEDATACLOUD_API_TOKEN": "<acedatacloud_api_token>",
"ACEDATACLOUD_API_BASE_URL": "https://..."
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall uvx mcp-grokversion 2026.8.28.0
What it does
MCP server for interacting with Grok AI models (chat, reasoning, vision) and generating videos from text or images via the Grok Imagine API. For developers who want to integrate Grok's chat capabilities and video generation into MCP-compatible applications.
- Generate short video clips from text descriptions using Grok Imagine video models.
- Animate still images into videos with optional motion prompts.
- Have multi-turn conversations with Grok reasoning models, including image analysis and tool calling.
Tools this server exposes
| Tool | Does |
|---|---|
| grok_chat_completions | Chat completion with Grok models, supporting reasoning, vision, and tool calling. |
| grok_text_to_video | Generate a video from a text prompt. |
| grok_image_to_video | Generate a video from an input image with optional motion prompt. |
| grok_get_task | Query the status and result of a single video generation task. |
| grok_get_tasks_batch | Query the status and results of multiple video generation tasks at once. |
| grok_list_models | List available Grok models and their capabilities. |
| grok_list_actions | List all available tools and example workflows. |
| grok_get_prompt_guide | Get tips for writing effective video generation prompts. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.