MySQL Mcp
Production safe MCP server that gives AI agents direct access to MySQL databases.
mysql-mcp
A Model Context Protocol (MCP) server that gives AI agents direct access to MySQL databases.
What It Can Do
Data Query & Mutation
execute_query— Run SELECT, INSERT, UPDATE, or DELETE statements with parameterized query support (?placeholders)execute_transaction— Execute multiple statements as an atomic batch; if any statement fails the entire transaction is automatically rolled back
Schema Management
create_table— Create a new tablealter_table— Add/drop columns, rename columns, add indexes, and moredrop_table— Drop a table (requiresALLOW_DESTRUCTIVE_DDL=true)show_tables— List all tables in a databasedescribe_table— Show column structure of a table
Database Management
list_databases— List all user databases (system databases excluded)create_database— Create a new database with optional charset and collationdrop_database— Drop a database (requiresALLOW_DESTRUCTIVE_DDL=true)
Introspection
ping— Check MySQL connection healthshow_columns— Show detailed column information for a tableget_server_info— Return MySQL version, character set, and collation (read-only)
Multi-Database Support
Every tool accepts a database parameter. You can query different databases within the same session without any extra configuration.
What It Cannot Do
The following operations are permanently blocked by the security filter and cannot be bypassed under any circumstances:
| Category | Blocked Commands |
|---|---|
| MySQL settings | SET GLOBAL, SET @@global.*, SET @@SESSION.sql_mode |
| User management | CREATE USER, DROP USER, ALTER USER, RENAME USER |
| Privilege management | GRANT, REVOKE |
| System commands | FLUSH, KILL, SHUTDOWN |
| Replication | RESET MASTER/REPLICA, START/STOP SLAVE/REPLICA, CHANGE MASTER |
| File access | LOAD DATA INFILE, SELECT INTO OUTFILE/DUMPFILE |
| Plugins | INSTALL PLUGIN, UNINSTALL PLUGIN |
Additional constraints:
- Multiple statements separated by
;in a single call are not allowed — useexecute_transactioninstead drop_tableanddrop_databaseare disabled by default and will return an error unlessALLOW_DESTRUCTIVE_DDL=trueis set- Query results are capped at
MAX_ROWSrows (default: 1000); add aLIMITclause for large tables
Installation
Requirements: Node.js 20+
git clone https://github.com/turkeryildirim/mysql-mcp
cd mysql-mcp
npm install
npm run build
Copy the example environment file and fill in your credentials:
cp .env.example .env
MYSQL_HOST=localhost
MYSQL_PORT=3306
MYSQL_USER=mcp_user
MYSQL_PASSWORD=your_password
# Optional
MYSQL_DATABASE=default_db
MYSQL_POOL_SIZE=10
MAX_ROWS=1000
ALLOW_DESTRUCTIVE_DDL=false
Creating a Dedicated MySQL User (Recommended)
Create a least-privilege MySQL user for the MCP server:
CREATE USER 'mcp_user'@'localhost' IDENTIFIED BY 'your_password';
GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, ALTER, INDEX
ON *.* TO 'mcp_user'@'localhost';
FLUSH PRIVILEGES;
Do not grant
SUPER,FILE,RELOAD, orGRANT OPTION.
Claude Desktop Integration
Open the Claude Desktop config file:
- Windows:
%APPDATA%\Claude\claude_desktop_config.json - macOS:
~/Library/Application Support/Claude/claude_desktop_config.json
Add the following block under mcpServers:
{
"mcpServers": {
"mysql": {
"command": "node",
"args": ["/absolute/path/to/mysql-mcp/dist/index.js"],
"env": {
"MYSQL_HOST": "localhost",
"MYSQL_PORT": "3306",
"MYSQL_USER": "mcp_user",
"MYSQL_PASSWORD": "your_password",
"MYSQL_DATABASE": "",
"MAX_ROWS": "1000",
"ALLOW_DESTRUCTIVE_DDL": "false"
}
}
}
}
Restart Claude Desktop. You should see the mysql server listed in the tools panel.
Claude CLI Integration
Option 1: .mcp.json — Project-scoped (Recommended)
Create a .mcp.json file in your project root:
{
"mcpServers": {
"mysql": {
"type": "stdio",
"command": "node",
"args": ["/absolute/path/to/mysql-mcp/dist/index.js"],
"env": {
"MYSQL_HOST": "localhost",
"MYSQL_PORT": "3306",
"MYSQL_USER": "mcp_user",
"MYSQL_PASSWORD": "your_password",
"ALLOW_DESTRUCTIVE_DDL": "false"
}
}
}
}
The server loads automatically whenever you run claude from that directory.
Option 2: Global configuration
claude mcp add mysql -- node /absolute/path/to/mysql-mcp/dist/index.js
To include environment variables:
claude mcp add mysql \
-e MYSQL_HOST=localhost \
-e MYSQL_USER=mcp_user \
-e MYSQL_PASSWORD=your_password \
-- node /absolute/path/to/mysql-mcp/dist/index.js
Verify the setup:
claude mcp list
claude mcp get mysql
Example Prompts
Once connected, you can instruct the agent in natural language:
Show me the structure of the users table in testdb
Fetch the last 10 orders from testdb.orders
Create a products table in testdb with id, name, price, and stock columns
Run the following two INSERTs as a single transaction: ...
Development
npm run dev # Watch mode (rebuilds on change)
npm test # Unit tests (no MySQL connection required)
npm run build # Production build → dist/
Máy chủ liên quan
OpenSearch MCP Server
An MCP server for interacting with OpenSearch clusters, configured via environment variables.
Kg
Lightweight knowledge graph
OceanBase
MCP Server for OceanBase database and its tools
Bamwor World Data
Access data on 261 countries and 13.4 million cities — population, GDP, geography, rankings, and comparisons. Built for Claude, Cursor, and AI agents.
SchemaFlow
Real-time PostgreSQL & Supabase database schema access for AI-IDEs via Model Context Protocol. Provides live database context through secure SSE connections with three powerful tools: get_schema, analyze_database, and check_schema_alignment.
AI Knowledge System
An enterprise-ready system to archive AI conversations from ChatGPT and Claude into a Supabase database.
NocoDB MCP Server
An MCP server for NocoDB, the open-source Airtable alternative. It allows interaction with your NocoDB instance via API.
MCP MySQL Server
An MCP server based on Spring AI that executes arbitrary SQL queries against a configured MySQL database.
CelestialMCP
Provides astronomical data such as celestial object positions, rise/set times, and visibility information.
Neo4j Knowledge Graph Memory
A knowledge graph memory server using the Neo4j graph database to store and retrieve information from AI interactions.