MCP Server MCP

Spark SQL

By github 👁 1 views ▲ 0 votes

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

Homepage Verified source GitHub

Install

uvx spark-sql-mcp-server

Configuration Example

{
  "remotes": [],
  "packages": [
    {
      "registryType": "pypi",
      "identifier": "spark-sql-mcp-server",
      "version": "0.1.2",
      "transport": {
        "type": "stdio"
      },
      "environmentVariables": [
        {
          "description": "Hostname of the Spark Thrift Server",
          "isRequired": true,
          "format": "string",
          "name": "SPARK_HOST"
        },
        {
          "description": "Port of the Spark Thrift Server (default: 10000)",
          "format": "number",
          "name": "SPARK_PORT"
        },
        {
          "description": "Default database to use",
          "format": "string",
          "name": "SPARK_DATABASE"
        },
        {
          "description": "Authentication method: NONE, LDAP, KERBEROS, CUSTOM, or NOSASL",
          "format": "string",
          "name": "SPARK_AUTH"
        },
        {
          "description": "Username for LDAP authentication",
          "format": "string",
          "name": "SPARK_USERNAME"
        },
        {
          "description": "Password for LDAP authentication",
          "format": "string",
          "isSecret": true,
          "name": "SPARK_PASSWORD"
        },
        {
          "description": "Kerberos service name (default: hive)",
          "format": "string",
          "name": "SPARK_KERBEROS_SERVICE_NAME"
        }
      ]
    }
  ]
}

mcp model-context-protocol pypi