Skip to content
VS Code

Spark SQL for VS Code

io.github.aidancorrell/spark-sql-mcp-server

Query Spark SQL clusters via Thrift/HiveServer2. Works with Spark, EMR, Hive, Impala.

client:VS Code transport:stdio runtime:pypi

Install Spark SQL in VS Code

.vscode/mcp.json

{
  "inputs": [
    {
      "type": "promptString",
      "id": "spark-host",
      "description": "SPARK_HOST",
      "password": true
    },
    {
      "type": "promptString",
      "id": "spark-port",
      "description": "SPARK_PORT",
      "password": true
    },
    {
      "type": "promptString",
      "id": "spark-database",
      "description": "SPARK_DATABASE",
      "password": true
    },
    {
      "type": "promptString",
      "id": "spark-auth",
      "description": "SPARK_AUTH",
      "password": true
    },
    {
      "type": "promptString",
      "id": "spark-username",
      "description": "SPARK_USERNAME",
      "password": true
    },
    {
      "type": "promptString",
      "id": "spark-password",
      "description": "SPARK_PASSWORD",
      "password": true
    },
    {
      "type": "promptString",
      "id": "spark-kerberos-service-name",
      "description": "SPARK_KERBEROS_SERVICE_NAME",
      "password": true
    }
  ],
  "servers": {
    "spark-sql-mcp-server": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "spark-sql-mcp-server"
      ],
      "env": {
        "SPARK_HOST": "${input:spark-host}",
        "SPARK_PORT": "${input:spark-port}",
        "SPARK_DATABASE": "${input:spark-database}",
        "SPARK_AUTH": "${input:spark-auth}",
        "SPARK_USERNAME": "${input:spark-username}",
        "SPARK_PASSWORD": "${input:spark-password}",
        "SPARK_KERBEROS_SERVICE_NAME": "${input:spark-kerberos-service-name}"
      }
    }
  }
}

VS Code prompts for each input the first time the server starts, so no secret is written to the file. VS Code docs

Spark SQL in other clients