katcoder-mysql-mcp

@berthojoris/katcoder-mysql-mcp

0 Stars 11 次浏览 berthojoris 更新于 2026-08-23

MCP 服务配置

复制以下 JSON 到 OPClaw 或其他 MCP 客户端的配置文件中即可使用

{
  "mcpServers": {
    "katkoder_mysql": {
      "args": [
        "-y",
        "katcoder-mysql-mcp",
        "mysql://root:password@localhost:3306/production_db",
        "all"
      ],
      "command": "npx"
    }
  }
}

服务介绍

KatCoder MySQL MCP Server

A secure and feature-rich MySQL Model Context Protocol (MCP) server that enables AI agents and applications to interact with MySQL databases through a standardized interface.

Features

Security First

SQL Injection Prevention: Comprehensive input validation and sanitization
Identifier Validation: Strict validation of table and column names
Query Whitelisting: Read-only operations by default, write operations require explicit permission
Connection Pooling: Secure connection management with timeout controls
Error Handling: Secure error messages that don't expose sensitive information

Database Operations

List: Browse tables and view table structures
Read: Query data with filtering, pagination, and sorting
Create: Insert new records with validation
Add Column: Add new columns to existing tables with full type and constraint support
Drop Column: Remove columns from tables with safety checks
Modify Column: Change column definitions (type, constraints, defaults)
Rename Column: Rename existing columns while preserving data
Rename Table: Rename tables with safety validation
Add Index: Create indexes (BTREE, HASH, FULLTEXT, SPATIAL) with unique constraints
Drop Index: Remove indexes from tables
Bulk Insert: Efficiently insert multiple records in a single operation
Update: Modify existing records safely
Delete: Remove records with mandatory WHERE clauses
Execute: Run custom SQL queries with security restrictions
DDL: Execute Data Definition Language statements
Transaction: Execute multiple operations atomically
Utility: Database health checks and metadata operations

Configuration Options

Connection String: Standard MySQL connection format
Tool Selection: Enable only the tools you need
Connection Pooling: Configurable pool settings
Timeout Controls: Connection and query timeouts

Tool Permissions & Security

Recommended Approach: Use "all" Tools

For most use cases, we recommend enabling all tools by using "all" as the tool parameter. This provides:

Full Functionality: Access to all database operations including DDL, transactions, and advanced features
AI Agent Compatibility: Ensures AI agents can see and use all available tools
Future-Proof: Automatically includes new tools as they're added
Simplified Configuration: No need to manually list specific tools

# Recommended: Enable all tools
npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/mydb" "all"

Security-First Approach: Manual Tool Selection

Use manual tool selection only when you need to restrict access for security or compliance reasons:

Read-Only Access

Perfect for reporting, analytics, or read-only AI agents:

npx katcoder-mysql-mcp "mysql://readonly:password@localhost:3306/mydb" "list,read,utility"

Available tools: list, read, utility

list: Browse tables and schema
read: Query data with filtering and pagination
utility: Database health checks and metadata

Basic Write Access

For applications that need to modify data but not schema:

npx katcoder-mysql-mcp "mysql://writer:password@localhost:3306/mydb" "list,read,create,update,delete,utility"

Available tools: list, read, create, update, delete, utility

Includes all read-only tools plus:
create: Insert new records
update: Modify existing records
delete: Remove records (with mandatory WHERE clauses)

Full Database Access

For database administrators and development environments:

npx katcoder-mysql-mcp "mysql://admin:password@localhost:3306/mydb" "all"

All available tools: list, read, create, update, delete, execute, ddl, transaction, bulk_insert, utility, add_column, drop_column, modify_column, rename_column, rename_table, add_index, drop_index, show_table_data

Security Considerations

Database User Permissions

Always use MySQL user accounts with appropriate privileges:

-- Read-only user
CREATE USER 'readonly'@'%' IDENTIFIED BY 'secure_password';
GRANT SELECT ON mydb.* TO 'readonly'@'%';

-- Write user (no DDL)
CREATE USER 'writer'@'%' IDENTIFIED BY 'secure_password';
GRANT SELECT, INSERT, UPDATE, DELETE ON mydb.* TO 'writer'@'%';

-- Admin user (full access)
CREATE USER 'admin'@'%' IDENTIFIED BY 'secure_password';
GRANT ALL PRIVILEGES ON mydb.* TO 'admin'@'%';

FLUSH PRIVILEGES;

Tool-Level vs Database-Level Security

Tool-level restrictions limit what operations the MCP server can perform
Database-level permissions provide the ultimate security boundary
Best practice: Use both layers for defense in depth

Production Recommendations

Use specific database users with minimal required privileges
Enable only necessary tools for production environments
Use read-only connections for reporting and analytics
Monitor database access and audit tool usage
Use environment variables for connection strings (never hardcode passwords)

Tool Selection Quick Reference

Use Case	Recommended Tools	Security Level
AI Development	`"all"`	Medium (use dev database)
Production AI	`"all"`	High (restricted DB user)
Reporting/Analytics	`"list,read,utility"`	High
Data Entry Apps	`"list,read,create,update,delete,utility"`	Medium
Database Admin	`"all"`	Low (trusted environment)
CI/CD Pipelines	`"all"`	Medium (isolated environment)

Installation

Note: This package is currently in development and not yet published to npm. Use the development installation method below.

Development Installation (Recommended)

git clone https://github.com/katkoder/katcoder-mysql-mcp.git
cd katcoder-mysql-mcp
npm install
npm run build

Future npm Installation (Coming Soon)

Once published to npm, you will be able to install globally:

# This will be available after publication
npm install -g katcoder-mysql-mcp

Local npm Installation (Coming Soon)

# This will be available after publication
npm install katcoder-mysql-mcp

Usage

Command Line Interface

Current Development Usage

# After building the project (npm run build)
# Basic usage with all tools enabled
node dist/cli.js "mysql://user:password@localhost:3306/database_name"

# With all tools enabled (recommended)
node dist/cli.js "mysql://user:password@localhost:3306/database_name" "all"

# With specific tools enabled (if you need to limit access)
node dist/cli.js "mysql://user:password@localhost:3306/database_name" "list,read,utility"

# With verbose logging
node dist/cli.js "mysql://user:password@localhost:3306/database_name" "all" --verbose

Future npm Usage (After Publication)

# Basic usage with all tools enabled
npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/database_name"

# With all tools enabled (recommended)
npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/database_name" "all"

# With specific tools enabled (if you need to limit access)
npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/database_name" "list,read,utility"

# With verbose logging
npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/database_name" "all" --verbose

Configuration for AI Agents

Current Development Configuration

Claude Desktop Configuration:
Add this configuration to your Claude Desktop configuration file:

{
  "mcpServers": {
    "katkoder_mysql": {
      "command": "node",
      "args": [
        "/path/to/katcoder-mysql-mcp/dist/cli.js",
        "mysql://root:password@localhost:3306/production_db",
        "all"
      ],
      "cwd": "/path/to/katcoder-mysql-mcp"
    }
  }
}

Cursor IDE Configuration:
For Cursor IDE, add to your settings:

{
  "mcp.servers": {
    "katkoder_mysql": {
      "command": "node",
      "args": [
        "/path/to/katcoder-mysql-mcp/dist/cli.js",
        "mysql://user:password@localhost:3306/development_db",
        "all"
      ],
      "cwd": "/path/to/katcoder-mysql-mcp"
    }
  }
}

Future npm Configuration (After Publication)

Claude Desktop Configuration:

{
  "mcpServers": {
    "katkoder_mysql": {
      "command": "npx",
      "args": [
        "-y",
        "katcoder-mysql-mcp",
        "mysql://root:password@localhost:3306/production_db",
        "all"
      ]
    }
  }
}

Cursor IDE Configuration:

{
  "mcp.servers": {
    "katkoder_mysql": {
      "command": "npx",
      "args": [
        "-y",
        "katcoder-mysql-mcp",
        "mysql://user:password@localhost:3306/development_db",
        "all"
      ]
    }
  }
}

Connection String Format

mysql://[user[:password]@]host[:port]/database

Basic Examples:

mysql://root@localhost:3306/mydb - Local database without password
mysql://user:password@localhost:3306/mydb - Local database with password
mysql://user:password@192.168.1.100:3306/mydb - Remote database

Advanced Examples:

mysql://user:password@db.example.com:3306/production?ssl=true - Remote database with SSL
mysql://root:password@mysql-container:3306/docker_db - Docker database
mysql://user:password@localhost:3307/alternative_port - Different port

Available Tools

1. List Tool

Browse database structure and table information.

Parameters:

table (optional): Specific table name to get column information

Examples:

{
  "name": "list",
  "arguments": {}
}

{
  "name": "list",
  "arguments": {
    "table": "users"
  }
}

Practical Usage Scenarios:

Database Discovery: When connecting to a new database, use the list tool without parameters to see all available tables
Schema Exploration: Use with a table name to understand the structure before writing queries
Data Modeling: Examine relationships between tables by checking foreign key constraints
Migration Planning: Understand existing schema before making changes

2. Read Tool

Query data from tables with filtering and pagination.

Parameters:

table (required): Table name to query
columns (optional): Array of specific columns to select
where (optional): Object with filter conditions
limit (optional): Maximum number of rows (max: 10,000)
offset (optional): Number of rows to skip
orderBy (optional): Order by clause

Basic Examples:

{
  "name": "read",
  "arguments": {
    "table": "users",
    "columns": ["id", "name", "email"],
    "where": {"status": "active"},
    "limit": 10,
    "orderBy": "created_at DESC"
  }
}

{
  "name": "read",
  "arguments": {
    "table": "products",
    "where": {"category": "electronics", "price": {"$gt": 100}},
    "limit": 50
  }
}

Advanced Filtering Examples:

{
  "name": "read",
  "arguments": {
    "table": "users",
    "columns": ["id", "email", "created_at"],
    "where": {"status": "active", "created_at": {"$gte": "2024-01-01"}},
    "limit": 25,
    "offset": 50,
    "orderBy": "last_login DESC"
  }
}

3. Bulk Insert Tool

Efficiently insert multiple records into a table in a single operation.

Parameters:

table (required): Target table name
data (required): Array of objects with identical column-value pairs

Examples:

{
  "name": "bulk_insert",
  "arguments": {
    "table": "users",
    "data": [
      {
        "name": "John Doe",
        "email": "john@example.com",
        "age": 30,
        "status": "active"
      },
      {
        "name": "Jane Smith",
        "email": "jane@example.com",
        "age": 25,
        "status": "active"
      },
      {
        "name": "Bob Wilson",
        "email": "bob@example.com",
        "age": 35,
        "status": "inactive"
      }
    ]
  }
}

Usage in Transactions:

{
  "name": "transaction",
  "arguments": {
    "operations": [
      {
        "type": "bulk_insert",
        "table": "users",
        "data": [
          {
            "name": "Alice Brown",
            "email": "alice@example.com",
            "age": 28,
            "status": "active"
          }
        ]
      },
      {
        "type": "update",
        "table": "user_stats",
        "data": { "total_users": 1 },
        "where": { "id": 1 }
      }
    ]
  }
}

Response Format:

{
  "success": true,
  "table": "users",
  "recordCount": 3,
  "affectedRows": 3,
  "insertedId": 1,
  "message": "Successfully inserted 3 records into users"
}

4. Create Tool

Insert new records into tables.

Parameters:

table (required): Target table name
data (required): Object with column-value pairs

Examples:

{
  "name": "create",
  "arguments": {
    "table": "users",
    "data": {
      "name": "John Doe",
      "email": "john@example.com",
      "status": "active"
    }
  }
}

4. Update Tool

Modify existing records safely.

Parameters:

table (required): Target table name
data (required): Object with column-value pairs to update
where (required): Object with filter conditions

Examples:

{
  "name": "update",
  "arguments": {
    "table": "users",
    "data": {
      "status": "inactive",
      "updated_at": "2024-01-01 12:00:00"
    },
    "where": {"id": 123}
  }
}

5. Delete Tool

Remove records with mandatory WHERE clauses.

Parameters:

table (required): Target table name
where (required): Object with filter conditions

Examples:

{
  "name": "delete",
  "arguments": {
    "table": "sessions",
    "where": {"expired": true}
  }
}

6. Execute Tool

Run custom SQL queries with security restrictions.

Parameters:

query (required): SQL query string
params (optional): Array of query parameters
allowWrite (optional): Boolean to allow write operations

Basic Examples:

{
  "name": "execute",
  "arguments": {
    "query": "SELECT COUNT(*) as total FROM users WHERE created_at > ?",
    "params": ["2024-01-01"]
  }
}

{
  "name": "execute",
  "arguments": {
    "query": "UPDATE users SET last_login = NOW() WHERE id = ?",
    "params": [123],
    "allowWrite": true
  }
}

Complex Query Examples:

{
  "name": "execute",
  "arguments": {
    "query": "SELECT u.email, COUNT(o.id) as order_count FROM users u LEFT JOIN orders o ON u.id = o.user_id GROUP BY u.id HAVING order_count > 5"
  }
}

{
  "name": "execute",
  "arguments": {
    "query": "SELECT DATE(created_at) as date, COUNT(*) as daily_signups FROM users WHERE created_at >= DATE_SUB(NOW(), INTERVAL 30 DAY) GROUP BY DATE(created_at) ORDER BY date",
    "params": []
  }
}

7. DDL Tool

Execute Data Definition Language statements.

Parameters:

statement (required): DDL statement

Examples:

{
  "name": "ddl",
  "arguments": {
    "statement": "CREATE INDEX idx_email ON users(email)"
  }
}

8. Add Column Tool

Add new columns to existing tables with comprehensive type and constraint support.

Parameters:

table (required): Target table name
column (required): Object with column definition
- name (required): New column name
- type (required): Column data type (e.g., VARCHAR(255), INT, DATETIME)
- nullable (optional): Whether column can contain NULL values
- default (optional): Default value for the column
- autoIncrement (optional): Whether column should auto-increment
- comment (optional): Column comment
position (optional): Object specifying column position
- after (optional): Place column after this existing column
- first (optional): Place column as the first column

Examples:

{
  "name": "add_column",
  "arguments": {
    "table": "users",
    "column": {
      "name": "email",
      "type": "VARCHAR(255)",
      "nullable": false,
      "default": "no-email@example.com"
    },
    "position": {
      "after": "name"
    }
  }
}

{
  "name": "add_column",
  "arguments": {
    "table": "products",
    "column": {
      "name": "is_active",
      "type": "BOOLEAN",
      "default": true,
      "comment": "Product availability status"
    }
  }
}

9. Drop Column Tool

Remove columns from tables with safety validation.

Parameters:

table (required): Table name to remove column from
column (required): Column name to drop

Examples:

{
  "name": "drop_column",
  "arguments": {
    "table": "users",
    "column": "old_field"
  }
}

10. Modify Column Tool

Change existing column definitions including type, constraints, and defaults.

Parameters:

table (required): Table name containing the column
column (required): Column name to modify
newDefinition (required): Object with new column definition
- type (required): New column data type
- nullable (optional): Whether column can contain NULL values
- default (optional): New default value
- comment (optional): Column comment

Examples:

{
  "name": "modify_column",
  "arguments": {
    "table": "users",
    "column": "age",
    "newDefinition": {
      "type": "INT",
      "nullable": true,
      "default": null
    }
  }
}

11. Rename Column Tool

Rename existing columns while preserving data.

Parameters:

table (required): Table name containing the column
oldName (required): Current column name
newName (required): New column name
newDefinition (optional): Column definition for the renamed column

Examples:

{
  "name": "rename_column",
  "arguments": {
    "table": "users",
    "oldName": "user_name",
    "newName": "username"
  }
}

12. Rename Table Tool

Rename tables with safety validation.

Parameters:

oldName (required): Current table name
newName (required): New table name

Examples:

{
  "name": "rename_table",
  "arguments": {
    "oldName": "user_profiles",
    "newName": "user_settings"
  }
}

13. Add Index Tool

Create indexes on tables for improved query performance.

Parameters:

table (required): Table name to add index to
name (required): Index name
columns (required): Array of column names to include in the index
type (optional): Index type (BTREE, HASH, FULLTEXT, SPATIAL)
unique (optional): Whether the index should be unique

Examples:

{
  "name": "add_index",
  "arguments": {
    "table": "users",
    "name": "idx_email",
    "columns": ["email"],
    "unique": true
  }
}

{
  "name": "add_index",
  "arguments": {
    "table": "products",
    "name": "idx_category_price",
    "columns": ["category_id", "price"],
    "type": "BTREE"
  }
}

14. Drop Index Tool

Remove indexes from tables.

Parameters:

table (required): Table name containing the index
name (required): Index name to drop

Examples:

{
  "name": "drop_index",
  "arguments": {
    "table": "users",
    "name": "idx_temp"
  }
}

15. Transaction Tool

Execute multiple operations atomically.

Parameters:

operations (required): Array of operations to execute in transaction

Basic Examples:

{
  "name": "transaction",
  "arguments": [
    {
      "type": "create",
      "table": "orders",
      "data": {"user_id": 123, "total": 99.99}
    },
    {
      "type": "update",
      "table": "users",
      "data": {"last_order_date": "2024-01-01"},
      "where": {"id": 123}
    }
  ]
}

Advanced Transaction Examples with Schema Changes:

{
  "name": "transaction",
  "arguments": {
    "operations": [
      {
        "type": "add_column",
        "table": "users",
        "column": {
          "name": "phone",
          "type": "VARCHAR(20)",
          "nullable": true
        }
      },
      {
        "type": "add_index",
        "table": "users",
        "name": "idx_phone",
        "columns": ["phone"],
        "unique": true
      },
      {
        "type": "update",
        "table": "users",
        "data": {"phone": "+1234567890"},
        "where": {"id": 1}
      }
    ]
  }
}

Response Format:

{
  "success": true,
  "operations": 3,
  "results": [
    {
      "description": "Add column 'phone' to table 'users'",
### v1.1.0 (Latest)
- **New Feature**: Added Comprehensive Schema Modification Tools
  - Implemented `add_column` tool for adding new columns with full type and constraint support
  - Implemented `drop_column` tool for safely removing columns from tables
  - Implemented `modify_column` tool for changing column definitions
  - Implemented `rename_column` tool for renaming existing columns
  - Implemented `rename_table` tool for renaming tables
  - Implemented `add_index` tool for creating various types of indexes
  - Implemented `drop_index` tool for removing indexes from tables
  - Added comprehensive schema validation and security measures
  - Enhanced transaction support for schema operations with rollback mechanisms
  - Added detailed documentation with examples and usage scenarios

### v1.0.1
- **New Feature**: Added Bulk Insert Tool for efficient multi-record insertion
  - Implemented `bulk_insert` tool for batch data imports
  - Supports inserting multiple records in a single database operation
  - Includes comprehensive validation and error handling
  - Can be used within transactions for atomic operations
  - Added detailed documentation with examples and usage scenarios

### v1.0.0
- Initial release
- All database operations implemented
- Comprehensive security features
- Full documentation
      "affectedRows": 0
    },
    {
      "description": "Create unique index 'idx_phone' on table 'users'",
      "affectedRows": 0
    },
    {
      "description": "Update user record with phone number",
      "affectedRows": 1
    }
  ]
}

{
"name": "ddl",
"arguments": {
"statement": "ALTER TABLE users ADD COLUMN phone VARCHAR(20)"
}
}


### 8. Transaction Tool
Execute multiple operations atomically.

**Parameters:**
- `operations` (required): Array of operations to execute

**Basic Examples:**
```json
{
  "name": "transaction",
  "arguments": {
    "operations": [
      {
        "type": "create",
        "table": "orders",
        "data": {"user_id": 123, "total": 99.99}
      },
      {
        "type": "update",
        "table": "users",
        "data": {"last_order_date": "2024-01-01"},
        "where": {"id": 123}
      }
    ]
  }
}

Advanced Transaction Examples:

{
  "name": "transaction",
  "arguments": {
    "operations": [
      {
        "type": "create",
        "table": "orders",
        "data": {"user_id": 123, "total": 99.99, "status": "pending"}
      },
      {
        "type": "update",
        "table": "users",
        "data": {"last_order_date": "2024-01-01"},
        "where": {"id": 123}
      },
      {
        "type": "create",
        "table": "order_items",
        "data": {"order_id": "LAST_INSERT_ID()", "product_id": 456, "quantity": 2}
      }
    ]
  }
}

9. Utility Tool

Database health checks and metadata operations.

Parameters:

action (required): Utility action (ping, version, stats, describe_table)
table (optional): Table name (required for describe_table)

Examples:

{
  "name": "utility",
  "arguments": {
    "action": "ping"
  }
}

{
  "name": "utility",
  "arguments": {
    "action": "stats"
  }
}

{
  "name": "utility",
  "arguments": {
    "action": "describe_table",
    "table": "users"
  }
}

10. Show Table Data Tool

Display table data with advanced formatting, pagination, and schema information.

Parameters:

table (required): Table name to display data from
limit (optional): Maximum number of rows to display (default: 50, max: 1000)
offset (optional): Number of rows to skip for pagination (default: 0)
columns (optional): Array of specific columns to display (default: all columns)
where (optional): Object with filter conditions (same format as read tool)
orderBy (optional): Column name to sort by (defaults to primary key or first column)
orderDirection (optional): Sort direction - 'ASC' or 'DESC' (default: 'ASC')
showSchema (optional): Include table schema information (default: true)
format (optional): Output format - 'table', 'json', or 'csv' (default: 'table')

Basic Examples:

{
  "name": "show_table_data",
  "arguments": {
    "table": "users"
  }
}

{
  "name": "show_table_data",
  "arguments": {
    "table": "products",
    "limit": 25,
    "columns": ["id", "name", "price", "category"],
    "orderBy": "price",
    "orderDirection": "DESC"
  }
}

Advanced Examples:

{
  "name": "show_table_data",
  "arguments": {
    "table": "orders",
    "where": {"status": "pending", "created_at": {"$gte": "2024-01-01"}},
    "limit": 100,
    "offset": 50,
    "format": "csv",
    "showSchema": false
  }
}

{
  "name": "show_table_data",
  "arguments": {
    "table": "users",
    "columns": ["id", "email", "last_login"],
    "where": {"status": "active"},
    "orderBy": "last_login",
    "orderDirection": "DESC",
    "format": "json"
  }
}

Response Format:

{
  "success": true,
  "table": "users",
  "format": "table",
  "pagination": {
    "currentPage": 1,
    "totalPages": 5,
    "limit": 50,
    "offset": 0,
    "totalRows": 247,
    "hasMore": true,
    "showing": "1-50 of 247"
  },
  "data": [...],
  "displayInfo": "formatted table string (for table format)",
  "count": 50,
  "schema": {
    "columns": [...],
    "totalColumns": 8
  }
}

Practical Usage Scenarios:

Data Exploration: Quickly browse table contents with automatic formatting
Data Export: Export table data in CSV format for external analysis
Debugging: View specific rows with filtering and pagination
Schema Analysis: Examine table structure alongside data
Report Generation: Generate formatted data displays for documentation

Security Features

SQL Injection Prevention

Input Sanitization: All table and column names are sanitized
Parameter Binding: All queries use parameterized statements
Query Validation: Dangerous SQL patterns are blocked
Write Operation Protection: Write operations require explicit permission

Identifier Validation

Table Names: Only alphanumeric characters and underscores allowed
Column Names: Validated against SQL injection patterns
Where Conditions: Values are checked for dangerous content

Connection Security

Connection Pooling: Secure connection management
Timeout Controls: Prevents hanging connections
Error Handling: Secure error messages without sensitive data

Security Best Practices

1. Recommended Setup: All Tools with Restricted Database User

Best practice for production and AI agents:

-- Create user with appropriate database-level permissions
CREATE USER 'mcp_ai_agent'@'localhost' IDENTIFIED BY 'secure_password';
GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, ALTER, INDEX, DROP ON myapp.* TO 'mcp_ai_agent'@'localhost';
FLUSH PRIVILEGES;

# Enable all tools - database permissions provide the security boundary
npx katcoder-mysql-mcp "mysql://mcp_ai_agent:secure_password@localhost:3306/myapp" "all"

Why this approach works:

AI agents can see and use all available tools
Database user permissions control actual access
Future-proof as new tools are automatically available
Simplified configuration management

2. Security-First Scenarios

Read-Only Analytics/Reporting

CREATE USER 'mcp_readonly'@'localhost' IDENTIFIED BY 'secure_password';
GRANT SELECT ON myapp.* TO 'mcp_readonly'@'localhost';
FLUSH PRIVILEGES;

# Restrict tools to read-only operations
npx katcoder-mysql-mcp "mysql://mcp_readonly:secure_password@localhost:3306/myapp" "list,read,utility"

Data Entry Applications (No Schema Changes)

CREATE USER 'mcp_writer'@'localhost' IDENTIFIED BY 'secure_password';
GRANT SELECT, INSERT, UPDATE, DELETE ON myapp.* TO 'mcp_writer'@'localhost';
FLUSH PRIVILEGES;

# Allow data operations but restrict DDL tools
npx katcoder-mysql-mcp "mysql://mcp_writer:secure_password@localhost:3306/myapp" "list,read,create,update,delete,bulk_insert,utility"

3. Development Environment

# Development: Use all tools with admin user
npx katcoder-mysql-mcp "mysql://root:password@localhost:3306/dev_db" "all"

4. Environment Variables (Recommended)

# Set connection string as environment variable
export MYSQL_URL="mysql://mcp_ai_agent:secure_password@localhost:3306/myapp"

# Use with all tools enabled
npx katcoder-mysql-mcp "$MYSQL_URL" "all"

# Or with specific tools for restricted access
npx katcoder-mysql-mcp "$MYSQL_URL" "list,read,utility"

5. Docker/Container Environments

# Using Docker secrets or environment variables
export MYSQL_URL="mysql://mcp_user:${DB_PASSWORD}@mysql-container:3306/production_db"
npx katcoder-mysql-mcp "$MYSQL_URL" "all"

Quick Reference: Choosing the Right Permission Approach

Use Case	Recommended Tools	Database Permissions	Security Level
AI Development & Prototyping	`"all"`	Full admin access	Low (dev only)
Production AI Agent	`"all"`	Limited to specific database/schema	High
Read-only Analytics	`"list,read,utility"`	SELECT only	High
Data Entry App	`"list,read,create,update,delete,bulk_insert,utility"`	No DDL permissions	Medium
Schema Migration Tool	`"all"`	DDL permissions required	Medium
Reporting Dashboard	`"list,read,utility"`	SELECT only	High

Most Common Setup (Recommended)

# 1. Create restricted database user
CREATE USER 'mcp_agent'@'localhost' IDENTIFIED BY 'secure_password';
GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, ALTER, INDEX, DROP ON your_app.* TO 'mcp_agent'@'localhost';

# 2. Use all tools - security handled by database permissions
npx katcoder-mysql-mcp "mysql://mcp_agent:secure_password@localhost:3306/your_app" "all"

Why this works: Database permissions provide the real security boundary, while "all" tools ensure AI agents can see and use all available functionality.

Error Handling

The server provides detailed error messages while maintaining security:

{
  "error": true,
  "message": "Table 'nonexistent_table' does not exist",
  "details": "Check the table name and try again"
}

Development

Building the Project

npm run build

Running in Development Mode

npm run dev

Testing

npm test

Environment Variables

LOG_LEVEL: Set logging level (debug, info, warn, error)
NODE_ENV: Set environment (development, production)

Troubleshooting

Connection Issues

Verify MySQL server is running
Check connection string format
Ensure database exists
Verify user permissions

Test Connection

# Test with all tools enabled
npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/mydb" "all"

# Then use: {"name": "utility", "arguments": {"action": "ping"}}

Check Database Version

npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/mydb" "all"

# Then use: {"name": "utility", "arguments": {"action": "version"}}

Permission Errors

Check MySQL user privileges
Ensure database access is granted
Verify table-level permissions

Performance Issues

Monitor connection pool usage
Check query execution times
Optimize database indexes

Monitor Performance

npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/mydb" "all"

# Then use: {"name": "utility", "arguments": {"action": "stats"}}

Advanced Configuration

Custom Connection Pool Settings

# Environment variables for connection tuning
export MYSQL_CONNECTION_LIMIT=20
export MYSQL_ACQUIRE_TIMEOUT=30000
export MYSQL_TIMEOUT=45000

npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/mydb"

Logging Configuration

# Enable debug logging
export LOG_LEVEL=debug

# Enable verbose output
npx katcoder-mysql-mcp "mysql://user:password@localhost:3306/mydb" "all" --verbose

Contributing

Fork the repository
Create a feature branch
Make your changes
Add tests
Submit a pull request

License

MIT License - see LICENSE file for details.

Support

For

查看源码项目主页