MySQL MCP Server
A Python MCP server for inspecting and querying MySQL databases, providing table discovery, schema inspection, read-only queries, and optional write/DDL tools.
README
<!-- mcp-name: io.github.musaddiq-dev/mysql-mcp-server -->
MySQL MCP Server
A Python Model Context Protocol (MCP) server for inspecting and querying MySQL databases from MCP-compatible clients. It provides table discovery, schema inspection, read query execution, DDL lookup, query explanation, and optional write/DDL tools for controlled database administration workflows.
Features
- List tables and describe table schemas
- Execute SELECT queries with safety checks
- Retrieve
SHOW CREATE TABLEoutput - Explain query execution plans
- Summarize tables and row counts
- Optional write and DDL tools for users who intentionally run with elevated database privileges
Safety Model
mysql_execute_read_query only accepts a single statement beginning with SELECT, rejects common modifying SQL keywords and risky file-read/write forms, and caps returned rows by MYSQL_READ_QUERY_LIMIT. This is a guardrail, not a substitute for database permissions. Use a dedicated read-only MySQL user for safe exploration. The write and DDL tools can modify or destroy data if the configured database user is allowed to do so; keep them on manual approval in your MCP client.
Requirements
- Python 3.11+
- MySQL 5.7+ or MySQL 8.0+
- MCP-compatible client such as Claude Desktop, Cursor, VS Code, or another MCP host
Installation
When published to PyPI, install or run the server like a standard Python MCP package:
uvx mdev-mysql-mcp-server
For local development from source:
git clone https://github.com/musaddiq-dev/mysql-mcp-server.git
cd mysql-mcp-server
python -m venv .venv
source .venv/bin/activate
pip install -e .
Configuration
Copy the example environment file and update it with your database connection details.
cp .env.example .env
| Variable | Description | Required | Default |
|---|---|---|---|
MYSQL_HOST |
MySQL host | No | localhost |
MYSQL_PORT |
MySQL port | No | 3306 |
MYSQL_USER |
MySQL username | No | root |
MYSQL_PASSWORD |
MySQL password | No | Empty |
MYSQL_DATABASE |
MySQL database name | Yes | Empty |
MYSQL_POOL_SIZE |
Connection pool size | No | 5 |
LOG_LEVEL |
Python logging level | No | INFO |
MYSQL_READ_QUERY_LIMIT |
Maximum rows returned by read queries | No | 1000 |
Example read-only user:
CREATE USER 'mcp_readonly'@'localhost' IDENTIFIED BY 'change-me';
GRANT SELECT ON your_database.* TO 'mcp_readonly'@'localhost';
FLUSH PRIVILEGES;
Running
mdev-mysql-mcp-server
From a local checkout before PyPI publication, run:
python -m mysql_mcp_server.server
MCP Client Configuration
For published installs, prefer uvx. MCP servers using stdio must write protocol messages only to stdout; this server writes logs to stderr through Python logging.
Claude Desktop / Cursor / Windsurf / Cline
Most MCP clients accept this mcpServers JSON shape:
{
"mcpServers": {
"mysql": {
"command": "uvx",
"args": ["mdev-mysql-mcp-server"],
"env": {
"MYSQL_HOST": "localhost",
"MYSQL_PORT": "3306",
"MYSQL_USER": "mcp_readonly",
"MYSQL_PASSWORD": "change-me",
"MYSQL_DATABASE": "your_database"
}
}
}
}
For local development from this repository, use the installed console script path instead:
{
"mcpServers": {
"mysql": {
"command": "/absolute/path/to/mysql-mcp-server/.venv/bin/mdev-mysql-mcp-server",
"args": [],
"env": {
"MYSQL_HOST": "localhost",
"MYSQL_PORT": "3306",
"MYSQL_USER": "mcp_readonly",
"MYSQL_PASSWORD": "change-me",
"MYSQL_DATABASE": "your_database"
}
}
}
}
Claude Code CLI
claude mcp add mysql \
--env MYSQL_HOST=localhost \
--env MYSQL_PORT=3306 \
--env MYSQL_USER=mcp_readonly \
--env MYSQL_PASSWORD=change-me \
--env MYSQL_DATABASE=your_database \
-- uvx mdev-mysql-mcp-server
VS Code MCP
VS Code uses the same command/args/env model in its MCP configuration:
{
"servers": {
"mysql": {
"type": "stdio",
"command": "uvx",
"args": ["mdev-mysql-mcp-server"],
"env": {
"MYSQL_HOST": "localhost",
"MYSQL_PORT": "3306",
"MYSQL_USER": "mcp_readonly",
"MYSQL_PASSWORD": "change-me",
"MYSQL_DATABASE": "your_database"
}
}
}
}
Tools
| Tool | Purpose | Safety |
|---|---|---|
mysql_list_tables |
List tables in the configured database | Read-only |
mysql_describe_table |
Show schema for a table | Read-only |
mysql_execute_read_query |
Execute a single bounded SELECT query | Read-only guardrail |
mysql_execute_write_query |
Execute a single INSERT, UPDATE, or DELETE | Destructive |
mysql_execute_ddl |
Execute a single CREATE, DROP, ALTER, or TRUNCATE | Destructive |
mysql_get_table_ddl |
Return SHOW CREATE TABLE output |
Read-only |
mysql_explain_query |
Run EXPLAIN for a single query |
Read-only |
mysql_get_database_summary |
Return table list and row counts | Read-only |
Smoke Check
Without a database, verify syntax with:
python -m py_compile src/mysql_mcp_server/server.py
With a configured database, start the server and use your MCP client to call list_tables.
Distribution
This server is published through the standard Python MCP distribution path:
- PyPI package:
mdev-mysql-mcp-server - MCP Registry name:
io.github.musaddiq-dev/mysql-mcp-server - Runtime hint:
uvx - Transport:
stdio
The mcp-name marker at the top of this README is required for MCP Registry ownership verification. Users should prefer uvx mdev-mysql-mcp-server in local MCP client configurations.
Security Notes
- Do not commit
.envor MCP client configs containing credentials. - Use least-privilege database users.
- Keep
execute_write_queryandexecute_ddlon explicit manual approval. - Do not expose this server over an untrusted network without additional authentication and transport security.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。