iunera/druid-mcp-server

Overview

This server implements the Model Context Protocol (MCP) to connect AI assistants with Apache Druid clusters. It exposes tools, resources, and AI-assisted prompts for managing and analyzing Druid, using Spring Boot and Spring AI. The server standardizes how AI agents interact with Druid through the MCP protocol.

Best for

Best for
Developers who want to connect AI assistants with Apache Druid using a standardized protocol

Use cases

• Query Druid datasets via natural language through an AI assistant
• Manage cluster configurations and schemas using chat interfaces
• Automate routine Druid analysis tasks with AI-driven prompts

How to use

Install

java -jar target/druid-mcp-server-2.0.0.jar

Tools exposed

• getDatasources
• getLookups
• getSegments

Tested with

ChatGPT, Claude, Gemini, Ypipe

Example client config

{\n  "servers": [\n    {\n      "name": "druid-mcp-server",\n      "url": "http://localhost:8080/mcp",\n      "transport": "http",\n      "auth": {\n        "type": "oauth2",\n        "token_url": "YOUR_TOKEN_URL",\n        "client_id": "YOUR_CLIENT_ID",\n        "client_secret": "YOUR_CLIENT_SECRET"\n      }\n    }\n  ]\n}

Notes

This server implements the Model Context Protocol (MCP) to connect AI assistants with Apache Druid clusters. It exposes tools, resources, and AI-assisted prompts for managing and analyzing Druid, using Spring Boot and Spring AI. The server standardizes how AI agents interact with Druid through the MCP protocol.

13 stars on GitHub. Last updated 2026-05-21. Licensed Apache-2.0.

Use cases

• Query Druid datasets via natural language through an AI assistant
• Manage cluster configurations and schemas using chat interfaces
• Automate routine Druid analysis tasks with AI-driven prompts

Pros

• Follows the open MCP standard for flexible AI tool integration
• Provides a comprehensive set of Druid-specific tools and resources
• Built on Spring Boot, offering a mature and stable Java framework

Cons

• Low GitHub star count (13) suggests limited community adoption
• Requires Java and Spring Boot environment, adding deployment overhead
• MCP protocol is less widely supported than REST or gRPC

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.