Alibaba Cloud ACK

Create a RAM sub-account and attach a policy granting read permissions for Container Service (cs), Log Service (log), and ARMS Prometheus. Then grant the sub-account RBAC access to your ACK clusters via the ACK console.

The recommended way to run ack-mcp-server is with uvx, which handles the Python environment automatically. Set your Alibaba Cloud credentials as environment variables before running.

export ACCESS_KEY_ID=your-access-key-id
export ACCESS_KEY_SECRET=your-access-key-secret
uvx alibabacloud-ack-mcp-server@latest

Register the server in your MCP client configuration file. The example below works for Claude Desktop (claude_desktop_config.json) and Claude Code.

{
  "mcpServers": {
    "ack-mcp-server": {
      "command": "uvx",
      "args": ["alibabacloud-ack-mcp-server@latest"],
      "env": {
        "ACCESS_KEY_ID": "your-access-key-id",
        "ACCESS_KEY_SECRET": "your-access-key-secret"
      }
    }
  }
}

For server-side or CI/CD deployments, run the official Docker image with your credentials passed as environment variables.

docker run -d \
  --name ack-mcp-server \
  -e ACCESS_KEY_ID="your-access-key-id" \
  -e ACCESS_KEY_SECRET="your-access-key-secret" \
  -p 8000:8000 \
  registry-cn-beijing.ack.aliyuncs.com/acs/ack-mcp-server:latest \
  python -m main_server --transport sse --host 0.0.0.0 --port 8000

Ask your AI assistant to list your ACK clusters. If credentials and permissions are correct, you will see your clusters returned. You can then query Prometheus metrics, run kubectl operations, and trigger diagnostics.