Apache Druid MCP server

This is a helper that connects your AI assistant to Apache Druid, a system many companies use to store and search through large amounts of data. Once it is set up, you can ask your AI questions about your Druid data or ask it to check on the health of your Druid setup, without learning any special query language. It is handy for data analysts, data engineers, and anyone who works with a Druid cluster but would rather ask than click around.

What this MCP server does

You ask your AI something like which data sources exist in Druid, or how many rows are in a table. The AI passes that request to this helper, and the helper talks to your Druid cluster using Druid's own web addresses. Druid sends back the answer, and the helper hands it to your AI, which explains it to you in the chat. Depending on how it is set up, it can also do things like submit data loading jobs or change retention rules, not just read.

Click to zoom

What you can do with it

• List all data sources in your Druid cluster
• Run SQL queries against your Druid data
• Check the health and status of your cluster
• Look at ingestion tasks and their logs
• View and manage retention rules
• Manage streaming supervisors (pause, resume, stop)
• Submit a data ingestion job

Try asking your AI

• “What data sources do I have in Druid?”
• “Show me the schema of the sales data source”
• “How many rows were added yesterday in the events table?”
• “Is my Druid cluster healthy right now?”

What it gives back to you

You get back plain answers in the chat: lists of data sources, tables of query results, status summaries, or short explanations of what the cluster is doing. If you ask it to change something, it tells you what it did. Everything shows up as normal text, not as code or files.

Before you start

Who it's for

Data analysts, data engineers, and IT staff who work with an Apache Druid cluster and want to ask questions or manage it in plain English.