Spark SQL

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

stdio v0.1.2

Connection Configuration

Add this to your AI agent's MCP config file to connect.

Run in your terminal:

claude mcp add io.github.aidancorrell--spark-sql-mcp-server -- uvx

Transport

stdio

Authentication

None

Version

v0.1.2

Server Name

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

Last Updated

Feb 8, 2026

How to install and connect this MCP server.

Environment variables you must provide when running this server.

Variable	Description	Required
`SPARK_HOST`	Hostname of the Spark Thrift Server	Yes
`SPARK_PORT`	Port of the Spark Thrift Server (default: 10000)	Optional
`SPARK_DATABASE`	Default database to use	Optional
`SPARK_AUTH`	Authentication method: NONE, LDAP, KERBEROS, CUSTOM, or NOSASL	Optional
`SPARK_USERNAME`	Username for LDAP authentication	Optional
`SPARK_PASSWORD` secret	Password for LDAP authentication	Optional
`SPARK_KERBEROS_SERVICE_NAME`	Kerberos service name (default: hive)	Optional