Doris MCP Server

Name: Doris MCP Server
Author: apache

@apache

代码仓库主页

关于 Doris MCP Server

Apache Doris MCP Server

#doris #warehouse

基本信息

分类

其他

许可证

Apache-2.0

运行时

python

传输方式

stdio

发布者

apache

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "doris-mcp": {
      "command": "uv",
      "args": [
        "--project",
        "/path/to/your/doris-mcp-server",
        "run",
        "doris-mcp-server"
      ],
      "env": {
        "DB_HOST": "FE_HOST",
        "DB_PORT": "9030",
        "DB_USER": "root",
        "DB_PASSWORD": ""
      }
    },
    "doris-http": {
      "url": "http://127.0.0.1:3000/mcp"
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Doris MCP Server?

Doris MCP Server is a Python and FastAPI backend that implements the Model Context Protocol (MCP) to connect Apache Doris databases with LLMs. It enables natural-language-to-SQL (NL2SQL) conversion, query execution, metadata management, and enterprise-grade security features such as token-based authentication and multi-tenant access control.

How to use Doris MCP Server?

Install via pip install doris-mcp-server, then start in HTTP mode with database connection parameters (--db-host, --db-port, --db-user, --db-password) or in stdio mode for direct integration with MCP clients like Cursor. Configure environment variables as an alternative to command-line arguments. A web-based token management dashboard is available for enterprise administration.

Key features of Doris MCP Server

MCP protocol with HTTP (streamable) and stdio transports
Enterprise authentication: Token, JWT, OAuth, and role-based access
Hot‑reload configuration management with zero‑downtime updates
SQL security validation, injection protection, and data masking
Session caching and connection pooling (60% overhead reduction)
Multi-worker horizontal scaling with stateless architecture

Use cases of Doris MCP Server

Convert natural language questions into SQL queries against Apache Doris
Execute and explain SQL statements with performance profiling
Manage metadata across multi-catalog environments (Hive, MySQL, etc.)
Provide secure, multi-tenant data access with token-bound database configuration
Monitor and analyze Doris cluster health and query performance

FAQ from Doris MCP Server

What are the system requirements?

Python 3.12+ and an Apache Doris database connection (host, port, user, password, database).

How do I connect to a Doris database?

Provide connection details via command-line arguments (--db-host, --db-port, --db-user, --db-password) or environment variables (DORIS_HOST, DORIS_PORT, DORIS_USER, DORIS_PASSWORD).

What transport modes are supported?

Two modes: http for a web service (default) and stdio for direct integration with MCP clients like Cursor.

Is authentication available?

Yes. v0.6.0 introduces enterprise authentication with Token, JWT, and OAuth support, including a web-based token management dashboard restricted to localhost.

Where is the database configuration stored?

Database credentials can be provided at startup (CLI or environment variables) or bound to tokens for multi-tenant setups. Token configurations are persisted in tokens.json.