MySQL MCP Server

A fast, read-only MySQL Server for the Model Context Protocol (MCP) written in Go.

This project exposes safe MySQL introspection tools to Claude Desktop via MCP. Claude can explore databases, describe schemas, and execute controlled read-only SQL queries — ideal for secure development assistance, debugging, analytics, and schema documentation.

Features

Fully read-only (blocks all non-SELECT/SHOW/DESCRIBE/EXPLAIN)
Multi-DSN Support: Connect to multiple MySQL or MariaDB instances, switch via tool
MariaDB Support: Native compatibility with MariaDB 10.x and 11.x
Vector Search (MySQL 9.0+): Similarity search on vector columns
MCP tools:
- list_databases, list_tables, describe_table
- run_query (safe and row-limited)
- ping, server_info
- list_connections, use_connection (multi-DSN)
- vector_search, vector_info (MySQL 9.0+)
Supports MySQL 8.0, 8.4, 9.0+ and MariaDB 10.x, 11.x
Query timeouts, structured logging, audit logs
Single Go binary
Unit and integration tests (Testcontainers)
Native integration with Claude Desktop MCP

Installation

Homebrew (macOS/Linux)

brew install askdba/tap/mysql-mcp-server

Update local installation (after a new release):

brew update && brew upgrade mysql-mcp-server

Docker

docker pull ghcr.io/askdba/mysql-mcp-server:latest

Note: Docker image tags use the raw version number without a leading "v" (e.g., 1.5.0, not v1.5.0).

Download Binary

Download the latest release from GitHub Releases.

Available for:

macOS (Intel & Apple Silicon)
Linux (amd64 & arm64)
Windows (amd64)

Build from Source

git clone https://github.com/askdba/mysql-mcp-server.git
cd mysql-mcp-server
make build

Binary output: bin/mysql-mcp-server

Quickstart

Option A: Homebrew (macOS/Linux)

brew install askdba/tap/mysql-mcp-server
mysql-mcp-server --version

Then follow the client-specific setup in the Claude Desktop, Claude Code, or Cursor IDE sections below.

Tip: brew info askdba/tap/mysql-mcp-server shows a config reminder after install.

Option B: Build from Source

git clone https://github.com/askdba/mysql-mcp-server.git
cd mysql-mcp-server
make build

When running from a cloned repo you can also use the interactive setup script:

./scripts/quickstart.sh

This will test your MySQL connection, optionally create a read-only MCP user, and generate your Claude Desktop configuration.

Configuration

Environment variables:

Variable	Required	Default	Description
MYSQL_DSN	Yes	–	MySQL DSN
MYSQL_MAX_ROWS	No	200	Max rows returned
MYSQL_QUERY_TIMEOUT_SECONDS	No	30	Query timeout
MYSQL_MCP_EXTENDED	No	0	Enable extended tools (set to 1)
MYSQL_MCP_JSON_LOGS	No	0	Enable JSON structured logging (set to 1)
MYSQL_MCP_TOKEN_TRACKING	No	0	Enable estimated token usage tracking (set to 1)
MYSQL_MCP_TOKEN_MODEL	No	cl100k_base	Tokenizer encoding to use for estimation
MYSQL_MCP_AUDIT_LOG	No	–	Path to audit log file
MYSQL_MCP_VECTOR	No	0	Enable vector tools for MySQL 9.0+ (set to 1)
MYSQL_MCP_HTTP	No	0	Enable REST API mode (set to 1)
MYSQL_HTTP_PORT	No	9306	HTTP port for REST API mode
MYSQL_HTTP_RATE_LIMIT	No	0	Enable rate limiting for HTTP mode (set to 1)
MYSQL_HTTP_RATE_LIMIT_RPS	No	100	Rate limit: requests per second
MYSQL_HTTP_RATE_LIMIT_BURST	No	200	Rate limit: burst size
MYSQL_MAX_OPEN_CONNS	No	10	Max open database connections
MYSQL_MAX_IDLE_CONNS	No	5	Max idle database connections
MYSQL_CONN_MAX_LIFETIME_MINUTES	No	30	Connection max lifetime in minutes
MYSQL_CONN_MAX_IDLE_TIME_MINUTES	No	5	Max idle time before connection is closed
MYSQL_PING_TIMEOUT_SECONDS	No	5	Database ping/health check timeout
MYSQL_HTTP_REQUEST_TIMEOUT_SECONDS	No	60	HTTP request timeout in REST API mode
MYSQL_SSL	No	–	Enable SSL/TLS for connections (true, false, skip-verify, preferred)

SSL/TLS Configuration

Enable encrypted connections to MySQL servers:

SSL Value	Description
`true`	Enable TLS with certificate verification
`false`	Disable TLS (default)
`skip-verify`	Enable TLS without certificate verification (self-signed certs)
`preferred`	Maps to `skip-verify` (Go MySQL driver limitation)

Note: The Go MySQL driver doesn't support tls=preferred. When you specify preferred, it is automatically mapped to skip-verify to ensure TLS is enabled.

Environment variable:

# Enable SSL for all connections
export MYSQL_SSL="true"

# Or per-connection SSL
export MYSQL_DSN_1_SSL="skip-verify"

Config file:

connections:
  production:
    dsn: "user:pass@tcp(prod:3306)/db?parseTime=true"
    ssl: "true"

SSH Tunneling (Bastion Host)

Connect to MySQL through an SSH bastion when the database is not directly reachable:

Variable	Description
`MYSQL_SSH_HOST`	Bastion hostname
`MYSQL_SSH_USER`	SSH username
`MYSQL_SSH_KEY_PATH`	Path to private key file
`MYSQL_SSH_PORT`	SSH port (default 22)

Environment variables:

export MYSQL_SSH_HOST="bastion.example.com"
export MYSQL_SSH_USER="deploy"
export MYSQL_SSH_KEY_PATH="$HOME/.ssh/id_rsa"
export MYSQL_DSN="user:pass@tcp(mysql.internal:3306)/mydb?parseTime=true"

Config file:

connections:
  production:
    dsn: "user:pass@tcp(mysql.internal:3306)/mydb?parseTime=true"
    ssh:
      host: "bastion.example.com"
      user: "deploy"
      key_path: "~/.ssh/id_rsa"
      port: 22  # optional, default 22

Multi-DSN Configuration

Configure multiple MySQL connections using numbered environment variables:

# Default connection
export MYSQL_DSN="user:pass@tcp(localhost:3306)/db1?parseTime=true"

# Additional connections
export MYSQL_DSN_1="user:pass@tcp(prod-server:3306)/production?parseTime=true"
export MYSQL_DSN_1_NAME="production"
export MYSQL_DSN_1_DESC="Production database"
export MYSQL_DSN_1_SSL="true"  # Enable SSL for this connection

export MYSQL_DSN_2="user:pass@tcp(staging:3306)/staging?parseTime=true"
export MYSQL_DSN_2_NAME="staging"
export MYSQL_DSN_2_DESC="Staging database"

Or use JSON configuration:

export MYSQL_CONNECTIONS='[
  {"name": "production", "dsn": "user:pass@tcp(prod:3306)/db?parseTime=true", "description": "Production", "ssl": "true"},
  {"name": "staging", "dsn": "user:pass@tcp(staging:3306)/db?parseTime=true", "description": "Staging"}
]'

Configuration File

As an alternative to environment variables, you can use a YAML or JSON configuration file.

Config file search order:

--config /path/to/config.yaml (command line flag)
MYSQL_MCP_CONFIG environment variable
./mysql-mcp-server.yaml (current directory)
~/.config/mysql-mcp-server/config.yaml (user config)
/etc/mysql-mcp-server/config.yaml (system config)

Example config file (mysql-mcp-server.yaml):

# Database connections
connections:
  default:
    dsn: "user:pass@tcp(localhost:3306)/mydb?parseTime=true"
    description: "Local development database"
  production:
    dsn: "readonly:pass@tcp(prod:3306)/prod?parseTime=true"
    description: "Production (read-only)"
    ssl: "true"  # Enable TLS with certificate verification

# Query settings
query:
  max_rows: 200
  timeout_seconds: 30

# Connection pool
pool:
  max_open_conns: 10
  max_idle_conns: 5
  conn_max_lifetime_minutes: 30

# Features
features:
  extended_tools: true
  vector_tools: false

# HTTP/REST API (optional)
http:
  enabled: false
  port: 9306

Command line options:

# Use specific config file
mysql-mcp-server --config /path/to/config.yaml

# Validate config file
mysql-mcp-server --validate-config /path/to/config.yaml

# Print current configuration as YAML
mysql-mcp-server --print-config

# Silent mode: suppress INFO/WARN logs (only errors to stderr)
mysql-mcp-server --silent --config /path/to/config.yaml

# Daemon mode: run HTTP server in background (Unix; use with MYSQL_MCP_HTTP=1)
MYSQL_MCP_HTTP=1 mysql-mcp-server --daemon --config /path/to/config.yaml

Silent and daemon mode: Use -s / --silent to reduce log noise in production (INFO and WARN are suppressed; ERROR still goes to stderr). Use -d / --daemon to run the HTTP server detached in the background on Unix. For long-running services, use the example systemd unit or launchd plist. See Silent and daemon mode for details.

Priority: Environment variables override config file values, allowing:

Base configuration in file
Environment-specific overrides via env vars
Docker/K8s secret injection via env vars

See examples/config.yaml and examples/config.json for complete examples.

Example:

export MYSQL_DSN="root:password@tcp(127.0.0.1:3306)/mysql?parseTime=true"
export MYSQL_MAX_ROWS=200
export MYSQL_QUERY_TIMEOUT_SECONDS=30

Run:

make run

Version Information

Check the installed version:

mysql-mcp-server --version

Output:

mysql-mcp-server v1.6.0
  Build time: 2025-12-21T11:43:11Z
  Git commit: a1b2c3d

Claude Desktop Integration

Edit your Claude Desktop configuration file:

macOS:

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

Windows:

%APPDATA%\Claude\claude_desktop_config.json

Linux:

~/.config/Claude/claude_desktop_config.json

Add:

{
  "mcpServers": {
    "mysql": {
      "command": "mysql-mcp-server",
      "env": {
        "MYSQL_DSN": "user:password@tcp(127.0.0.1:3306)/mydb?parseTime=true",
        "MYSQL_MAX_ROWS": "200",
        "MYSQL_MCP_EXTENDED": "1"
      }
    }
  }
}

Note: If Claude Desktop cannot find the binary, replace "mysql-mcp-server" with the full path from which mysql-mcp-server.

Restart Claude Desktop.

Cursor IDE Integration

Cursor IDE supports the Model Context Protocol (MCP). Configure it to use mysql-mcp-server:

Edit your Cursor MCP configuration file:

macOS:

~/.cursor/mcp.json

Windows:

%APPDATA%\Cursor\mcp.json

Linux:

~/.config/Cursor/mcp.json

Add:

{
  "mcpServers": {
    "mysql": {
      "command": "mysql-mcp-server",
      "env": {
        "MYSQL_DSN": "root:password@tcp(127.0.0.1:3306)/mydb?parseTime=true",
        "MYSQL_MAX_ROWS": "200",
        "MYSQL_MCP_EXTENDED": "1"
      }
    }
  }
}

With Docker:

{
  "mcpServers": {
    "mysql": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-e", "MYSQL_DSN=root:password@tcp(host.docker.internal:3306)/mydb?parseTime=true",
        "-e", "MYSQL_MCP_EXTENDED=1",
        "ghcr.io/askdba/mysql-mcp-server:latest"
      ]
    }
  }
}

Restart Cursor after saving the configuration.

Claude Code Integration

Claude Code supports MCP servers via the CLI or a project-scoped .mcp.json file.

Option 1: CLI (quick setup)

claude mcp add --transport stdio \
  --env MYSQL_DSN="user:password@tcp(127.0.0.1:3306)/mydb?parseTime=true" \
  --env MYSQL_MCP_EXTENDED=1 \
  mysql -- mysql-mcp-server

Option 2: Project `.mcp.json`

This repo includes a .mcp.json that Claude Code auto-discovers when you open the project. Set your DSN in the shell before starting Claude Code:

export MYSQL_DSN="user:password@tcp(127.0.0.1:3306)/mydb?parseTime=true"
claude  # start Claude Code — MySQL tools will be available automatically

To use the same config in your own project, copy .mcp.json to your project root and set MYSQL_DSN in your shell.

Verify the integration by asking Claude Code: "List all databases on this MySQL server."

MCP Tools

list_databases

Returns non-system databases.

list_tables

Input:

{ "database": "employees" }

describe_table

Input:

{ "database": "employees", "table": "salaries" }

run_query

Input:

{ "sql": "SELECT id, name FROM users LIMIT 5" }

Optional database context:

{ "sql": "SELECT * FROM users LIMIT 5", "database": "myapp" }

Rejects non-read-only SQL
Enforces row limit
Enforces timeout

ping

Tests database connectivity and returns latency.

Output:

{ "success": true, "latency_ms": 2, "message": "pong" }

server_info

Returns MySQL server details.

Output:

{
  "version": "8.0.36",
  "version_comment": "MySQL Community Server - GPL",
  "uptime_seconds": 86400,
  "current_database": "myapp",
  "current_user": "mcp@localhost",
  "character_set": "utf8mb4",
  "collation": "utf8mb4_0900_ai_ci",
  "max_connections": 151,
  "threads_connected": 5
}

list_connections

List all configured MySQL connections.

Output:

{
  "connections": [
    {"name": "production", "dsn": "user:****@tcp(prod:3306)/db", "active": true},
    {"name": "staging", "dsn": "user:****@tcp(staging:3306)/db", "active": false}
  ],
  "active": "production"
}

use_connection

Switch to a different MySQL connection.

Input:

{ "name": "staging" }

Output:

{
  "success": true,
  "active": "staging",
  "message": "Switched to connection 'staging'",
  "database": "staging_db"
}

Vector Tools (MySQL 9.0+)

Enable with:

export MYSQL_MCP_VECTOR=1

vector_search

Perform similarity search on vector columns.

Input:

{
  "database": "myapp",
  "table": "embeddings",
  "column": "embedding",
  "query": [0.1, 0.2, 0.3, ...],
  "limit": 10,
  "select": "id, title, content",
  "distance_func": "cosine"
}

Output:

{
  "results": [
    {"distance": 0.123, "data": {"id": 1, "title": "Doc 1", "content": "..."}},
    {"distance": 0.456, "data": {"id": 2, "title": "Doc 2", "content": "..."}}
  ],
  "count": 2
}

Distance functions: cosine (default), euclidean, dot

vector_info

List vector columns in a database.

Input:

{ "database": "myapp" }

Output:

{
  "columns": [
    {"table": "embeddings", "column": "embedding", "dimensions": 768, "index_name": "vec_idx"}
  ],
  "vector_support": true,
  "mysql_version": "9.0.0"
}

Extended Tools (MYSQL_MCP_EXTENDED=1)

Enable with:

export MYSQL_MCP_EXTENDED=1

list_indexes

List indexes on a table.

{ "database": "myapp", "table": "users" }

show_create_table

Get the CREATE TABLE statement.

{ "database": "myapp", "table": "users" }

explain_query

Get execution plan for a SELECT query.

{ "sql": "SELECT * FROM users WHERE id = 1", "database": "myapp" }

list_views

List views in a database.

{ "database": "myapp" }

list_triggers

List triggers in a database.

{ "database": "myapp" }

list_procedures

List stored procedures.

{ "database": "myapp" }

list_functions

List stored functions.

{ "database": "myapp" }

list_partitions

List table partitions.

{ "database": "myapp", "table": "events" }

database_size

Get database size information.

{ "database": "myapp" }

Or get all databases:

{}

table_size

Get table size information.

{ "database": "myapp" }

foreign_keys

List foreign key constraints.

{ "database": "myapp", "table": "orders" }

list_status

List MySQL server status variables.

{ "pattern": "Threads%" }

list_variables

List MySQL server configuration variables.

{ "pattern": "%buffer%" }

Security Model

SQL Safety (Paranoid Mode)

The server enforces strict SQL validation:

Allowed operations:

SELECT, SHOW, DESCRIBE, EXPLAIN

Blocked patterns:

Multi-statement queries (semicolons)
File operations: LOAD_FILE(), INTO OUTFILE, INTO DUMPFILE
DDL: CREATE, ALTER, DROP, TRUNCATE, RENAME
DML: INSERT, UPDATE, DELETE, REPLACE
Admin: GRANT, REVOKE, FLUSH, KILL, SHUTDOWN
Dangerous functions: SLEEP(), BENCHMARK(), GET_LOCK()
Transaction control: BEGIN, COMMIT, ROLLBACK

Recommended MySQL User

CREATE USER 'mcp'@'localhost' IDENTIFIED BY 'strongpass';
GRANT SELECT ON *.* TO 'mcp'@'localhost';

Observability

JSON Structured Logging

Enable JSON logs for production:

export MYSQL_MCP_JSON_LOGS=1

Output:

{"timestamp":"2025-01-15T10:30:00.123Z","level":"INFO","message":"query executed","fields":{"tool":"run_query","duration_ms":15,"row_count":42}}

Audit Logging

Enable query audit trail:

export MYSQL_MCP_AUDIT_LOG=/var/log/mysql-mcp-audit.jsonl

Each query is logged with timing, success/failure, and row counts.

Token Usage Estimation (Optional)

Enable estimated token counting for tool inputs/outputs to monitor LLM context usage:

export MYSQL_MCP_TOKEN_TRACKING=1
export MYSQL_MCP_TOKEN_MODEL=cl100k_base  # default, used by GPT-4/Claude

Or via YAML config file:

logging:
  token_tracking: true
  token_model: "cl100k_base"

When enabled:

JSON logs include a tokens object with estimated input/output/total tokens
Audit log entries for run_query include input_tokens and output_tokens
All other tools also emit token estimates when token_tracking is enabled

Example JSON log output:

{
  "level": "INFO",
  "msg": "query executed",
  "tool": "run_query",
  "duration_ms": 45,
  "row_count": 10,
  "tokens": {
    "input_estimated": 25,
    "output_estimated": 150,
    "total_estimated": 175,
    "model": "cl100k_base"
  }
}

Notes:

Token counts are estimates using tiktoken encoding, not actual LLM billing
For payloads exceeding 1MB, a heuristic (~4 bytes per token) is used to prevent memory spikes
The feature is disabled by default to avoid overhead when not needed

Query Timing

All queries are automatically timed and logged with:

Execution duration (milliseconds)
Row count returned
Tool name
Truncated query (for debugging)

Performance Tuning

Connection Pool

Configure the connection pool for your workload:

export MYSQL_MAX_OPEN_CONNS=20      # Max open connections
export MYSQL_MAX_IDLE_CONNS=10      # Max idle connections  
export MYSQL_CONN_MAX_LIFETIME_MINUTES=60  # Connection lifetime

Testing

Prerequisites

Go 1.24+
Docker and Docker Compose (for integration tests)
MySQL client (optional, for manual database access)

Unit Tests

Run unit tests (no external dependencies):

make test

Integration Tests

Integration tests run against real MySQL instances using Docker Compose.

Test against MySQL 8.0 (default):

make test-integration-80

Test against MySQL 8.4:

make test-integration-84

Test against MySQL 9.0:

make test-integration-90

Test against all supported versions (MySQL & MariaDB):

make test-integration-all

Test specifically against MariaDB 11.4:

make test-integration-mariadb-11

Sakila Database Tests

The Sakila sample database provides comprehensive integration testing with a realistic DVD rental store schema featuring:

16 tables with foreign key relationships
Views, stored procedures, and triggers
FULLTEXT indexes and ENUM/SET types
Sample data for complex query testing

Run Sakila tests:

# Against MySQL 8.0
make test-sakila

# Against MySQL 8.4
make test-sakila-84

# Against MySQL 9.0
make test-sakila-90

For a multi-version Sakila test matrix (including alternate ports), see docs/sakila-test-steps.md.

The Sakila tests cover:

Multi-table JOINs (film→actors, customer→address→city→country)
Aggregation queries (COUNT, SUM, AVG, GROUP BY)
Subqueries (IN, NOT IN, correlated)
View queries
FULLTEXT search
Index and foreign key verification

Security Tests

Run SQL injection and validation security tests:

make test-security

Manual Database Management

Start and stop test MySQL containers manually:

# Start MySQL 8.0 container
make test-mysql-up

# Start all MySQL versions (8.0, 8.4, 9.0)
make test-mysql-all-up

# Stop all test containers
make test-mysql-down

# View container logs
make test-mysql-logs

Environment Variables

When running tests manually, set the DSN:

# MySQL 8.0 (port 3306)
export MYSQL_TEST_DSN="root:testpass@tcp(localhost:3306)/testdb?parseTime=true"

# MySQL 8.4 (port 3307)
export MYSQL_TEST_DSN="root:testpass@tcp(localhost:3307)/testdb?parseTime=true"

# MySQL 9.0 (port 3308)
export MYSQL_TEST_DSN="root:testpass@tcp(localhost:3308)/testdb?parseTime=true"

Then run tests directly:

go test -v -tags=integration ./tests/integration/...

Test Database Schema

Integration tests use two databases:

testdb (tests/sql/init.sql):

Tables: users, orders, products, special_data
Views: user_orders
Stored procedures: get_user_by_id
Sample test data for basic integration tests

sakila (tests/sql/sakila-schema.sql, tests/sql/sakila-data.sql):

16 tables: actor, film, customer, rental, payment, inventory, etc.
6 views: film_list, customer_list, staff_list, sales_by_store, etc.
Stored functions and procedures
FULLTEXT indexes on film_text
Realistic sample data (50 actors, 30 films, 10 customers, 20 rentals)

Docker

Using Pre-built Image

Basic usage:

docker run -e MYSQL_DSN="user:password@tcp(host.docker.internal:3306)/mydb" \
  ghcr.io/askdba/mysql-mcp-server:latest

Note: Use host.docker.internal instead of localhost to connect from inside the container to MySQL on your host machine (macOS/Windows).

With extended tools enabled:

docker run \
  -e MYSQL_DSN="user:password@tcp(host.docker.internal:3306)/mydb" \
  -e MYSQL_MCP_EXTENDED=1 \
  ghcr.io/askdba/mysql-mcp-server:latest

With all options:

docker run \
  -e MYSQL_DSN="user:password@tcp(host.docker.internal:3306)/mydb" \
  -e MYSQL_MCP_EXTENDED=1 \
  -e MYSQL_MCP_VECTOR=1 \
  -e MYSQL_MAX_ROWS=500 \
  -e MYSQL_QUERY_TIMEOUT_SECONDS=60 \
  ghcr.io/askdba/mysql-mcp-server:latest

Claude Desktop with Docker

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

{
  "mcpServers": {
    "mysql": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-e", "MYSQL_DSN=user:password@tcp(host.docker.internal:3306)/mydb",
        "ghcr.io/askdba/mysql-mcp-server:latest"
      ]
    }
  }
}

With extended tools:

{
  "mcpServers": {
    "mysql": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-e", "MYSQL_DSN=user:password@tcp(host.docker.internal:3306)/mydb",
        "-e", "MYSQL_MCP_EXTENDED=1",
        "ghcr.io/askdba/mysql-mcp-server:latest"
      ]
    }
  }
}

Docker Compose

services:
  mysql:
    image: mysql:8.4
    environment:
      MYSQL_ROOT_PASSWORD: rootpass
      MYSQL_DATABASE: testdb
    ports:
      - "3306:3306"

  mcp:
    image: ghcr.io/askdba/mysql-mcp-server:latest
    depends_on:
      - mysql
    environment:
      MYSQL_DSN: "root:rootpass@tcp(mysql:3306)/testdb?parseTime=true"
      MYSQL_MCP_EXTENDED: "1"

Run:

docker compose up

Build Locally

docker build -t mysql-mcp-server .

REST API Mode

Enable HTTP REST API mode to use with ChatGPT, Gemini, or any HTTP client:

export MYSQL_DSN="user:password@tcp(localhost:3306)/mydb"
export MYSQL_MCP_HTTP=1
export MYSQL_HTTP_PORT=9306  # Optional, defaults to 9306
mysql-mcp-server

Rate Limiting

Enable per-IP rate limiting for production deployments:

export MYSQL_HTTP_RATE_LIMIT=1
export MYSQL_HTTP_RATE_LIMIT_RPS=100    # 100 requests/second
export MYSQL_HTTP_RATE_LIMIT_BURST=200  # Allow bursts up to 200

When rate limited, clients receive HTTP 429 (Too Many Requests) with a Retry-After: 1 header.

Running as a service (daemon)

To run the REST API server in the background or under a process manager:

Unix (foreground detach): MYSQL_MCP_HTTP=1 mysql-mcp-server --daemon --config /path/to/config.yaml (forks and exits the parent; child runs detached).
systemd: Copy and customize contrib/systemd/mysql-mcp-server.service, then systemctl enable --now mysql-mcp-server.
macOS launchd: Copy and customize contrib/launchd/com.askdba.mysql-mcp-server.plist into ~/Library/LaunchAgents/ or /Library/LaunchDaemons/.
Windows: Use a Windows Service wrapper or run in a terminal; --daemon is not supported.

Use --silent to suppress INFO/WARN logs when running under a service manager. See docs/silent-and-daemon.md.

API Endpoints

Method	Endpoint	Description
GET	`/health`	Health check
GET	`/api`	API index with all endpoints
GET	`/api/databases`	List databases
GET	`/api/tables?database=`	List tables
GET	`/api/describe?database=&table=`	Describe table
POST	`/api/query`	Run SQL query
GET	`/api/ping`	Ping database
GET	`/api/server-info`	Server info
GET	`/api/connections`	List connections
POST	`/api/connections/use`	Switch connection

Extended endpoints (requires MYSQL_MCP_EXTENDED=1):

Method	Endpoint	Description
GET	`/api/indexes?database=&table=`	List indexes
GET	`/api/create-table?database=&table=`	Show CREATE TABLE
POST	`/api/explain`	Explain query
GET	`/api/views?database=`	List views
GET	`/api/triggers?database=`	List triggers
GET	`/api/procedures?database=`	List procedures
GET	`/api/functions?database=`	List functions
GET	`/api/partitions?database=&table=`	List table partitions
GET	`/api/size/database?database=`	Database size
GET	`/api/size/tables?database=`	Table sizes
GET	`/api/foreign-keys?database=`	Foreign keys
GET	`/api/status?pattern=`	Server status
GET	`/api/variables?pattern=`	Server variables

Vector endpoints (requires MYSQL_MCP_VECTOR=1):

Method	Endpoint	Description
POST	`/api/vector/search`	Vector similarity search
GET	`/api/vector/info?database=`	Vector column info

Example Usage

List databases:

curl http://localhost:9306/api/databases

Run a query:

curl -X POST http://localhost:9306/api/query \
  -H "Content-Type: application/json" \
  -d '{"sql": "SELECT * FROM users LIMIT 5", "database": "myapp"}'

Get server info:

curl http://localhost:9306/api/server-info

Response Format

All responses follow this format:

{
  "success": true,
  "data": { ... }
}

Error responses:

{
  "success": false,
  "error": "error message"
}

ChatGPT Custom GPT Integration

Start the REST API server on a publicly accessible host
Create a Custom GPT with Actions
Import the OpenAPI schema from /api
Configure authentication if needed

Docker with REST API

docker run -p 9306:9306 \
  -e MYSQL_DSN="user:password@tcp(host.docker.internal:3306)/mydb" \
  -e MYSQL_MCP_HTTP=1 \
  -e MYSQL_MCP_EXTENDED=1 \
  ghcr.io/askdba/mysql-mcp-server:latest

Documentation

SQL Query Optimization Guide: Practical optimization patterns and query rewriting techniques using the Stack Exchange schema.
Comprehensive MySQL Query Optimization Guide: Deep technical insights into the query optimizer, advanced indexing strategies, execution plan analysis, and operational best practices.

Project Structure

cmd/mysql-mcp-server/
├── main.go             -> Server entrypoint and tool registration
├── types.go            -> Input/output struct types for tools
├── tools.go            -> Core MCP tool handlers
├── tools_extended.go   -> Extended MCP tool handlers
├── http.go             -> HTTP REST API handlers and server
├── connection.go       -> Multi-DSN connection manager
└── logging.go          -> Structured and audit logging

internal/
├── api/                -> HTTP middleware and response utilities
├── config/             -> Configuration loader from environment
├── mysql/              -> MySQL client wrapper + tests
└── util/               -> Shared utilities (SQL validation, identifiers)

examples/               -> Example configs and test data
scripts/                -> Quickstart and utility scripts
bin/                    -> Built binaries

Examples

The examples/ folder contains:

claude_desktop_config.json - Example Claude Desktop configuration
test-dataset.sql - Demo database with tables, views, and sample data

Load the test dataset:

mysql -u root -p < examples/test-dataset.sql

This creates a mcp_demo database with:

5 categories, 13 products, 8 customers
9 orders with 16 order items
Views: order_summary, product_inventory
Stored procedure: GetCustomerOrders
Stored function: GetProductStock

Development

make fmt       # Format code
make lint      # Run linter
make test      # Run unit tests
make build     # Build binary
make release   # Build release binaries

Releasing

Releases are automated via GitHub Actions and GoReleaser. Full checklist: docs/releasing.md.

Quick steps: update CHANGELOG.md, commit, then git tag vX.Y.Z and git push origin vX.Y.Z. After the workflow runs, complete post-release steps (CHANGELOG on main if needed, Homebrew tap README, local brew upgrade). See the doc for details.

Name		Name	Last commit message	Last commit date
Latest commit History 164 Commits
.github		.github
cmd/mysql-mcp-server		cmd/mysql-mcp-server
contrib		contrib
docs		docs
examples		examples
homebrew		homebrew
internal		internal
scripts		scripts
tests		tests
.dockerignore		.dockerignore
.gitguardian.yaml		.gitguardian.yaml
.gitignore		.gitignore
.golangci.yml		.golangci.yml
.goreleaser.yml		.goreleaser.yml
.mcp.json		.mcp.json
CHANGELOG.md		CHANGELOG.md
Dockerfile		Dockerfile
LICENSE		LICENSE
Makefile		Makefile
MysqlMCPServerBanner.png		MysqlMCPServerBanner.png
README.md		README.md
docker-compose.test.yml		docker-compose.test.yml
go.mod		go.mod
go.sum		go.sum

Folders and files

Latest commit

History

Repository files navigation

MySQL MCP Server

Features

Installation

Homebrew (macOS/Linux)

Docker

Download Binary

Build from Source

Quickstart

Option A: Homebrew (macOS/Linux)

Option B: Build from Source

Configuration

SSL/TLS Configuration

SSH Tunneling (Bastion Host)

Multi-DSN Configuration

Configuration File

Version Information

Claude Desktop Integration

Cursor IDE Integration

Claude Code Integration

Option 1: CLI (quick setup)

Option 2: Project .mcp.json

MCP Tools

list_databases

list_tables

describe_table

run_query

ping

server_info

list_connections

use_connection

Vector Tools (MySQL 9.0+)

vector_search

vector_info

Extended Tools (MYSQL_MCP_EXTENDED=1)

list_indexes

show_create_table

explain_query

list_views

list_triggers

list_procedures

list_functions

list_partitions

database_size

table_size

foreign_keys

list_status

list_variables

Security Model

SQL Safety (Paranoid Mode)

Recommended MySQL User

Observability

JSON Structured Logging

Audit Logging

Token Usage Estimation (Optional)

Query Timing

Performance Tuning

Connection Pool

Testing

Prerequisites

Unit Tests

Integration Tests

Sakila Database Tests

Security Tests

Manual Database Management

Environment Variables

Test Database Schema

Docker

Using Pre-built Image

Claude Desktop with Docker

Docker Compose

Build Locally

REST API Mode

Rate Limiting

Running as a service (daemon)

API Endpoints

Example Usage

Option 2: Project `.mcp.json`

Packages