Skip to content

ClickHouse MCP Server

Query ClickHouse for real-time analytics. Execute SQL, inspect schemas, and analyze large-scale columnar data.

Capabilities
SQL Analytics Schema Columnar Data
Maintained by
ClickHouse
Hosting
Runs locally on your machine
Authentication
API key or credentials (environment variables)
Last checked

How to connect the ClickHouse MCP server

Pick your AI client and paste the snippet. Swap the YOUR_… placeholders for your own values. New to this? Our guide to how MCP clients and servers fit together explains what each piece does.

Claude Code

Terminal
claude mcp add --env CLICKHOUSE_HOST=YOUR_HOST --env CLICKHOUSE_USER=YOUR_USER --env CLICKHOUSE_PASSWORD=YOUR_PASSWORD --transport stdio clickhouse \
  -- uv run --with mcp-clickhouse --python 3.12 mcp-clickhouse
Full Claude Code setup guide →

Claude Desktop

claude_desktop_config.json (Settings → Developer → Edit Config)
{
  "mcpServers": {
    "clickhouse": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "mcp-clickhouse",
        "--python",
        "3.12",
        "mcp-clickhouse"
      ],
      "env": {
        "CLICKHOUSE_HOST": "YOUR_HOST",
        "CLICKHOUSE_USER": "YOUR_USER",
        "CLICKHOUSE_PASSWORD": "YOUR_PASSWORD"
      }
    }
  }
}

Fully quit and reopen Claude Desktop after saving.

Full Claude Desktop setup guide →

Cursor

~/.cursor/mcp.json (or .cursor/mcp.json in a project)
{
  "mcpServers": {
    "clickhouse": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "mcp-clickhouse",
        "--python",
        "3.12",
        "mcp-clickhouse"
      ],
      "env": {
        "CLICKHOUSE_HOST": "YOUR_HOST",
        "CLICKHOUSE_USER": "YOUR_USER",
        "CLICKHOUSE_PASSWORD": "YOUR_PASSWORD"
      }
    }
  }
}
Full Cursor setup guide →

VS Code

.vscode/mcp.json
{
  "servers": {
    "clickhouse": {
      "type": "stdio",
      "command": "uv",
      "args": [
        "run",
        "--with",
        "mcp-clickhouse",
        "--python",
        "3.12",
        "mcp-clickhouse"
      ],
      "env": {
        "CLICKHOUSE_HOST": "YOUR_HOST",
        "CLICKHOUSE_USER": "YOUR_USER",
        "CLICKHOUSE_PASSWORD": "YOUR_PASSWORD"
      }
    }
  }
}
Full VS Code setup guide →

Frequently asked questions

What is the ClickHouse MCP server?

It is a Model Context Protocol (MCP) server — a small piece of software that gives AI assistants like Claude a secure, governed connection to ClickHouse. Query ClickHouse for real-time analytics. Execute SQL, inspect schemas, and analyze large-scale columnar data.

Is there an official ClickHouse MCP server?

Yes. The ClickHouse MCP server is an official integration, maintained by the vendor or the Model Context Protocol project, which makes it the safest default for business use. It is maintained by ClickHouse.

Is the ClickHouse MCP server hosted, or do I run it myself?

It runs locally. Your AI client starts it on your own machine with uv, and it talks to the client over standard input and output rather than the network.

Do I need an API key for the ClickHouse MCP server?

Yes. The local server reads CLICKHOUSE_HOST, CLICKHOUSE_USER, CLICKHOUSE_PASSWORD from its environment, so you create those credentials first and add them to your client's config.

What can an AI assistant actually do with ClickHouse?

Through this server an assistant can work with sql, analytics, schema, columnar data — reading from and acting on ClickHouse directly instead of you copy-pasting between windows. What you allow it to do is controlled by the permissions you grant.

How do I set this up for my business?

The setup section above has copy-paste config for Claude Code, Claude Desktop, Cursor and VS Code. Once it is running against real systems, the question becomes what catches it when it acts wrongly, which is what Crox's Safety Diagnostic looks at.

New to MCP? Start with our plain-English guide to the Model Context Protocol or see how to connect AI to your business tools without writing code .

Done-for-you integration

Already running this in production?

An agent with a live connection to your tools is a model that can act. Crox looks at what catches it when it acts wrongly: the layers around it, who can stop it, and where a near miss gets reported. Two days, one team, one model.