JDBC-MCP - Database Access for AI Assistants

Name: JDBC-MCP - Database Access for AI Assistants
Author: guanghuang

@guanghuang

代码仓库

关于 JDBC-MCP - Database Access for AI Assistants

JDBC MCP modified from quarkus-mcp-servers

基本信息

分类

数据库

许可证

Apache-2.0 license

运行时

java

传输方式

stdio

发布者

guanghuang

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "jdbc-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-d",
        "--name",
        "jdbc-mcp",
        "\\"
      ]
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is JDBC-MCP - Database Access for AI Assistants?

JDBC-MCP is a Model Context Protocol (MCP) server that enables AI assistants to interact with databases through JDBC connections. It allows large language models (LLMs) to inspect, query, create, and modify database content. This fork of the Quarkus MCP JDBC Server adds Docker deployment support and enhanced security with write queries disabled by default.

How to use JDBC-MCP - Database Access for AI Assistants?

Run the server as a Docker container using either SSE (Streaming Server Events) or STDIO transport. Set required environment variables (jdbc.url, jdbc.user, jdbc.password) and optionally enable.write.sql. For STDIO with Claude Desktop, configure the Docker command in claude_desktop_config.json. An environment file (.env) can be used for configuration.

Key features of JDBC-MCP - Database Access for AI Assistants

Docker deployment via SSE or STDIO transport.
Write queries (INSERT, UPDATE, DELETE) disabled by default for security.
Supports multiple database types through JDBC.
Configurable via environment variables.
Compatible with Claude Desktop and other MCP clients.

Use cases of JDBC-MCP - Database Access for AI Assistants

AI assistants querying read-only database content.
Developers inspecting database schemas with a language model.
Securely modifying database records when explicitly enabled.
Automating database interactions from containerized environments.

FAQ from JDBC-MCP - Database Access for AI Assistants

How do I enable write operations?

Set the environment variable enable.write.sql=true when running the Docker container.

What database types are supported?

For details on supported JDBC variants, refer to the original Quarkus MCP JDBC Server documentation.

How can I restrict the server to read-only access?

Leave enable.write.sql unset or set to false (the default). Also use a read-only database user for additional security.

What transport modes are available?

SSE (HTTP) and STDIO. For SSE, the endpoint is http://localhost:8080/mcp/sse. STDIO requires the environment variables quarkus.mcp.server.stdio.enabled=true, quarkus.log.console.enable=false, and quarkus.log.console.stderr=false.

Where does the data live?

The server connects to an external database specified by the jdbc.url; data remains in that database. The server itself does not store data persistently.