Which integrations are available for this server?

Enables management of MariaDB databases, including capabilities for schema inspection, SQL query execution, and data migration. Provides tools for interacting with MySQL databases, allowing for schema exploration, SQL query execution, and synchronization across different environments. Allows for the management of PostgreSQL instances, including schema comparison, data migration, and the execution of SQL commands. Facilitates interaction with SQLite databases, supporting schema discovery, SQL queries, and database replication or cloning.

How do I use DBA MCP Server?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@DBA MCP Server Compare the staging and production schemas and generate a migration script." That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

DBA MCP Server 🚀

Multi-vendor Database Tool for DevOps, Migrations & Knowledge Management

dba-mcp is a Model Context Protocol (MCP) server designed to enable LLMs (like Claude, ChatGPT, or local models) to interact with, manage, and migrate SQL databases directly. It operates over stdio, making it ideal for integration with tools like Claude Desktop, Cursor, or any MCP client.

It combines "Infrastructure as Code" principles with persistent knowledge management, allowing AI agents to save specific context about databases (Governance, Business Rules, Schema Explanations) and recall it on demand.

🌟 Features

Multi-Vendor Support: Connect to Oracle, PostgreSQL, MySQL/MariaDB, SQL Server, and SQLite simultaneously.
Hybrid Configuration:
- Stateless: Inject connections via CLI arguments (--[id] [connection_string]).
- Persistent: Save connections permanently using the save_connection tool (stored in local SQLite).
Context & Knowledge Management 🧠:
- Save AI-generated context (Markdown) about a database (e.g., "This endpoint creates users based on table X").
- Recall context automatically to answer complex questions about specific databases.
Schema Intelligence:
- Compare schemas between different engines.
- Generate DDL scripts (CREATE, ALTER) to synchronize structures.
Data Migration:
- migrate_data: Move data between connections with optimized bulk inserts.
- replicate_database: Auto-Magic cloning of an entire database (Schema + Data).

📦 Installation

Via NPX (Recommended)

You can run it directly without installing. Connections defined in CLI are loaded alongside any persistent connections saved previously.

npx @grec0/dba-mcp --local sqlite://data.db --prod postgres://user:pass@host/db

Local Installation

git clone https://github.com/grecoLab/mcp-dba.git cd mcp-dba npm install npm run build

🛠️ Configuration & Persistence

1. CLI Arguments (Ephemeral/Stateless)

Ideal for CI/CD or temporary sessions. Format: --[id] [connection_string]

/* claude_desktop_config.json */ { "mcpServers": { "dba-tools": { "command": "npx", "args": [ "-y", "@grec0/dba-mcp", "--staging", "mysql://root:pass@localhost:3306/staging_db" ] } } }

2. Persistent Storage (Stateful)

You can use tools to save connections. They are stored in an internal SQLite database (.dba-mcp/storage.db) within the running directory.

Save: save_connection
Remove: remove_connection
List: list_connections (Shows source: 💾=Saved, 🖥️=CLI)

🧰 Available Tools for LLMs

Once connected, the LLM will have access to these tools:

Connection & Context Management

Tool	Description
`list_connections`	Lists all active connections. Icons indicate status: 💾 Saved to Disk 🖥️ From CLI Args 📝 Context Available
`save_connection`	Saves a new connection URL permanently.
`remove_connection`	Removes a stored connection.
`save_database_context`	Saves AI-generated markdown documentation for a specific connection (Governance, Business Logic).
`get_database_context`	Retrieves the stored markdown context for a connection.

Database Operations

Tool	Description
`inspect_schema`	Explore the database structure. Pass `tableName` to see columns/PKs, or empty to list all tables.
`sql_query`	Execute a read-only SQL query on a specific connection.
`generate_migration_ddl`	Compares two schemas (Source -> Target) and generates the SQL DDL commands to make Target match Source.
`migrate_data`	Migrates specific tables from Source to Target using bulk inserts.
`replicate_database`	🔥 Power Tool: Clones an ENTIRE database. Scans source, creates tables in target, and migrates all data automatically.

🧠 Supported Connection Strings

SQLite: sqlite://path/to/file.db
PostgreSQL: postgres://user:pass@host:5432/db
MySQL: mysql://user:pass@host:3306/db
Oracle: oracle://user:pass@host:1521/service
SQL Server: sqlserver://user:pass@host:1433/db

💻 Development & Release

Build

npm run build

This runs TypeScript compiler and ensures the binary is executable (shx chmod +x).

Release

To version and publish (follows Semantic Versioning):

npm run release:patch # 0.0.1 -> 0.0.2 npm run release:minor # 0.1.0 -> 0.2.0 npm run release:major # 1.0.0 -> 2.0.0

Note: These scripts automatically push to Git and publish to NPM.

Inspector

To debug the MCP server communication visually:

npm run inspector -- --local sqlite://test.db

🐳 Docker

A multi-stage Dockerfile is included for containerized deployments.

docker build -t dba-mcp . docker run -i dba-mcp --prod postgres://...

License

MIT

This server cannot be installed

-

security - not tested

F

license - not found

-

quality - not tested

How are these scores calculated?

Resources

GitHub Repository

Need Help?

Report Issue

Related Servers

DBA MCP Server