返回 MCP 目录
public公开dns本地运行

GreptimeDB AI服务

一种实现了模型上下文协议的服务器,使人工智能助手能够安全地与GreptimeDB交互,允许它们通过受控接口探索数据库模式、读取数据和执行SQL查询。

article

README

greptimedb-mcp-server

PyPI - Version build workflow MIT License

A Model Context Protocol (MCP) server implementation for GreptimeDB.

This server provides AI assistants with a secure and structured way to explore and analyze databases. It enables them to list tables, read data, and execute SQL queries through a controlled interface, ensuring responsible database access.

Project Status

This is an experimental project that is still under development. Data security and privacy issues have not been specifically addressed, so please use it with caution.

Capabilities

  • list_resources to list tables
  • read_resource to read table data
  • list_tools to list tools
  • call_tool to execute an SQL
  • list_prompts to list prompts
  • get_prompt to get the prompt by name

Installation

pip install greptimedb-mcp-server

Configuration

Set the following environment variables:

GREPTIMEDB_HOST=localhost    # Database host
GREPTIMEDB_PORT=4002         # Optional: Database MySQL port (defaults to 4002 if not specified)
GREPTIMEDB_USER=root
GREPTIMEDB_PASSWORD=
GREPTIMEDB_DATABASE=public
GREPTIMEDB_TIMEZONE=UTC

Or via command-line args:

  • --host the database host, localhost by default,
  • --port the database port, must be MySQL protocol port, 4002 by default,
  • --user the database username, empty by default,
  • --password the database password, empty by default,
  • --database the database name, public by default.
  • --timezone the session time zone, empty by default(using server default time zone).

Usage

Claude Desktop Integration

Configure the MCP server in Claude Desktop's configuration file:

MacOS

Location: ~/Library/Application Support/Claude/claude_desktop_config.json

Windows

Location: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "greptimedb": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/greptimedb-mcp-server",
        "run",
        "-m",
        "greptimedb_mcp_server.server"
      ],
      "env": {
        "GREPTIMEDB_HOST": "localhost",
        "GREPTIMEDB_PORT": "4002",
        "GREPTIMEDB_USER": "root",
        "GREPTIMEDB_PASSWORD": "",
        "GREPTIMEDB_DATABASE": "public",
        "GREPTIMEDB_TIMEZONE": ""
      }
    }
  }
}

License

MIT License - see LICENSE.md file for details.

Contribute

Prerequisites

  • Python with uv package manager
  • GreptimeDB installation
  • MCP server dependencies

Development

# Clone the repository
git clone https://github.com/GreptimeTeam/greptimedb-mcp-server.git
cd greptimedb-mcp-server

# Create virtual environment
uv venv
source venv/bin/activate  # or `venv\Scripts\activate` on Windows

# Install development dependencies
uv sync

# Run tests
pytest

Use MCP Inspector for debugging:

npx @modelcontextprotocol/inspector uv \
  --directory \
  /path/to/greptimedb-mcp-server \
  run \
  -m \
  greptimedb_mcp_server.server

Acknowledgement

This library's implementation was inspired by the following two repositories and incorporates their code, for which we express our gratitude:

Thanks!

help

运行方式说明

cloud

托管运行

托管运行通常表示这个 MCP Server 由服务方环境承载,用户一般按页面提供的连接方式或授权流程接入,不需要在本地长期启动一个 MCP 进程

  1. 打开服务方连接页
  2. 完成授权或复制端点
  3. 在 MCP 客户端中连接
terminal

本地运行 / 其它方式

本地运行通常需要用户在自己的电脑或服务器上安装依赖,把 server_config 复制到 MCP 客户端,并按 env_schema 补齐环境变量、密钥或其它配置

  1. 复制 server_config
  2. 安装所需依赖
  3. 补齐环境变量后重启客户端