Direct Database Access MCP Server
Enables chat-based interaction with PostgreSQL databases, supporting table listing, queries, write operations, and saved parameterized queries, deployable as a remote MCP server via Cloudflare Workers.
README
About
- MCP SERVER where allow you chat with your database(Postgresql) using AI
- You can deploy this MCP server as a remote MCP server using Cloudflare Workers
Key features
- List the tables of your database
- Query the tables of your database
- Execute write operations like INSERT/UPDATE/DELETE (privileged users only)
- Create queries and saved to execute later. PS: you can create queries with parameters and save the queries to execute later.
- All operations sent to Cloudflare Workers will check the api-key sent in the header
Folder structure
- src/
- database/ // Database connection and utils
- tools/ // Tools implementation
- repositories/ // Repositories implementation
Transport Protocols
This MCP server supports both modern and legacy transport protocols:
/mcp- Streamable HTTP (recommended): Uses a single endpoint with bidirectional communication, automatic connection upgrades, and better resilience for network interruptions/sse- Server-Sent Events (legacy): Uses separate endpoints for requests/responses, maintained for backward compatibility
For new implementations, use the /mcp endpoint as it provides better performance and reliability.
Prerequisites
- Node.js installed on your machine
- A Cloudflare account (free tier works)
- A PostgreSQL database (local or hosted)
Step 1: Install Wrangler CLI
Install Wrangler globally to manage your Cloudflare Workers:
npm install -g wrangler
Step 2: Authenticate with Cloudflare
Log in to your Cloudflare account:
wrangler login
This will open a browser window where you can authenticate with your Cloudflare account.
Step 3: Clone and Install dependencies
Clone the repo directly & install dependencies: pnpm install.
Environment Variables Setup
Before running the MCP server, you need to configure several environment variables for authentication and database access.
Create Environment Variables File
-
Create your
.dev.varsfile from the example:cp .dev.vars.example .dev.vars -
Configure all required environment variables in
.dev.vars:ACCESS_TOKEN=random_value_to_use_as_api_key DATABASE_URL=postgresql://username:password@localhost:5432/database_name
Database setup
- To save the queries to execute later you will need execute the SQL script on your database. PS: you need to execute the script named instructions.sql only once.
Run the server locally
wrangler dev
This makes the server available at http://localhost:8792
Production Deployment
Deploy
Deploy the MCP server to make it available on your workers.dev domain
wrangler deploy
Access the remote MCP server from Kilo Code and Vscode
You can access the remote MCP server from Kilo Code and Vscode using the following configuration:
// Kilocode MCP
{
"mcpServers": {
"direct-database-mcp-server-prod": {
"url": "url_cloudflare_workers_generated_after_deployed/sse",
"headers": {
"api-key": "Bearer same_value_of_ACCESS_TOKEN"
}
}
}
}
// Vscode MCP
{
"servers": {
"direct-database-mcp-server-prod": {
"url": "https://my-mcp-server.tiagorosadacost.workers.dev/sse",
"headers": {
"api-key": "Bearer 701cdccb-e605-423f-a778-66ed0698117f"
}
}
}
}
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。