Druid MCP Server
About
An MCP server for comprehensive read-only access to one or more Apache Druid clusters for ad-hoc data analysis, monitoring, troubleshooting, and comparison.
Explore
- Multi-Cluster Support: Connect to and query multiple Druid clusters simultaneously
- Query Execution: SQL and native JSON queries
- Datasource Management: List, explore, and inspect datasource schemas
- Ingestion Monitoring: Supervisor and task status tracking
- Cluster Operations: Service health monitoring and segment analysis
- Lookup Management: Query and inspect lookup tables
- Smart Analysis Prompts: Pre-built prompts for common data analysis tasks
- Schema Caching: Efficient caching with 5-minute TTL for better performance in resources only
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Druid MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
- Python 3.11+
- Access to an Apache Druid cluster
- MCP-compatible client (e.g., Claude Desktop)
```python
dev_count = execute_sql_query("dev", "SELECT COUNT() FROM orders")
prod_count = execute_sql_query("prod", "SELECT COUNT() FROM orders")
for cluster in ["dev", "stage", "prod"]:
status = get_cluster_status(cluster)
print(f"{cluster}: {status}")
The MCP server exposes 18 tools organized by functionality:
Important: All tools (except cluster management tools) require an explicit cluster parameter as the first argument. This enables efficient multi-cluster operations without state management.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"druid mcp server": {
"druid": {
"command": "uv",
"args": [
"--directory",
"/ABSOLUTE/PATH/TO/druid-mcp",
"run",
"main.py"
],
"env": {
"DRUID_CLUSTERS": "dev=http://localhost:8088 stage=http://localhost:8089 prod=http://localhost:8090"
}
}
}
}
}McpServers
{
"druid": {
"command": "uv",
"args": [
"--directory",
"/ABSOLUTE/PATH/TO/druid-mcp",
"run",
"main.py"
],
"env": {
"DRUID_CLUSTERS": "dev=http://localhost:8088 stage=http://localhost:8089 prod=http://localhost:8090"
}
}
}This repository provides an MCP server for read-only access to one or more Apache Druid clusters.
This is particularly useful for ad-hoc data analysis and comparison across Druid clusters.
For example, if you have multiple environments, each with their own Druid cluster, but ingesting data from the same
origin, then you could use a simple prompt like this to run a complex cross-cluster comparison:
> Check and compare the datasource for SumUp across all Druid clusters.
The AI agent should then list the Druid clusters, list their datasources, explore their schemas, segments, tasks, and
their data distribution all by itself, and ultimately present you a nice summary.
Overview
The Druid MCP server enables AI applications to interact with Apache Druid through the Model Context Protocol (MCP). It provides a standardized interface for querying and exploring Druid datasources, making it easy to integrate Druid data into AI workflows.
Features
- Multi-Cluster Support: Connect to and query multiple Druid clusters simultaneously
- Query Execution: SQL and native JSON queries
- Datasource Management: List, explore, and inspect datasource schemas
- Ingestion Monitoring: Supervisor and task status tracking
- Cluster Operations: Service health monitoring and segment analysis
- Lookup Management: Query and inspect lookup tables
- Smart Analysis Prompts: Pre-built prompts for common data analysis tasks
- Schema Caching: Efficient caching with 5-minute TTL for better performance in resources only
Prerequisites
- Python 3.11+
- Access to an Apache Druid cluster
- MCP-compatible client (e.g., Claude Desktop)
Installation
Install dependencies using uv:
uv install
Or with pip:
pip install -e .
Configuration
Multi-Cluster Support
Configure multiple Druid clusters with whitespace-separated key=value pairs:
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
