Multi-DB MCP Server
A multi-database MCP server supporting PostgreSQL, ClickHouse, and MySQL that enables database exploration and SQL execution through MCP stdio or HTTP API.
README
ai_mcp: Multi-DB MCP Server
一个可直接部署的多数据库 MCP 服务,支持 PostgreSQL / ClickHouse / MySQL。仓库结构和体验目标参考 mysql_mcp_server:
- 一次拉取即可使用
.env配置驱动- 既支持 MCP stdio,也支持 HTTP API 调用
目录
db_mcp_service/server.py: MCP stdio 工具(list_databases/list_tables/describe_table/execute_sql)db_mcp_service/api.py: FastAPI 对外 HTTP 接口db_mcp_service/run_api.py: HTTP 启动入口requirements.txt: 运行依赖pyproject.toml: 可选的包安装元信息(含命令入口)start-mcp.ps1/.cmd,start-api.ps1/.cmd: 一键启动脚本Dockerfile,docker-compose.yml: 一键容器启动examples/claude-desktop.json: Claude Desktop 配置示例
第一步:拉取并准备环境
git clone <your-repo-url>
cd ai_mcp
python -m venv .venv
. .venv\Scripts\Activate.ps1
python -m pip install -r requirements.txt
第二步:配置
copy .env.example .env
编辑 .env,只开启需要的后端,例如只开 PostgreSQL:
POSTGRES_ENABLED=true
POSTGRES_DSN=postgresql://user:password@127.0.0.1:5432/postgres
CLICKHOUSE_ENABLED=false
MYSQL_ENABLED=false
启动方式(两条主线)
1) MCP stdio(推荐给 MCP 客户端)
# 推荐直接用项目内脚本
./start-mcp.ps1
# 或直接运行模块
python -m db_mcp_service
2) HTTP API(给第三方系统调用)
./start-api.ps1
# 或
python -m db_mcp_service.run_api
HTTP 默认监听:0.0.0.0:8000,可在 .env 中改:
DB_MCP_API_HOSTDB_MCP_API_PORTDB_MCP_API_RELOAD
可选:安装成全局命令
如果你的环境可以安装 build 依赖,可执行:
python -m pip install -e .
安装后可直接使用:
db-mcp(MCP stdio)db-mcp-api(FastAPI)
如果该方式不可用,直接用上面的 python -m 启动方式即可。两者行为一致。
HTTP API 路由
GET /healthGET /databasesGET /databases/{database}GET /databases/{database}/tables?schema=publicGET /databases/{database}/tables/{table}?schema=publicPOST /sql
POST /sql 示例:
{
"database": "postgres",
"sql": "SELECT 1",
"params": null,
"allow_write": false
}
默认只读。allow_write=true 或设置全局 MCP_ALLOW_WRITE=true 后才能执行写语句。
API 鉴权(可选)
.env 配置:
DB_MCP_API_TOKEN=your-token
设置后访问 /sql 需要携带 Header:X-API-Key: your-token。
Docker 启动(可直接给别人用)
docker compose up --build
直接可复用的 Claude Desktop 配置
见文件:examples/claude-desktop.json
你可以把它复制到 Claude Desktop 的配置中,仅需改数据库连接字段即可直接共享。
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.