MCP.so
登录

MCPg : Production Grade Postgresql Mcp Server

@devopam

关于 MCPg : Production Grade Postgresql Mcp Server

A safe-by-default PostgreSQL Model Context Protocol server for AI agents.

基本信息

分类

数据库

传输方式

stdio

发布者

devopam

提交者

Devopam Mittra

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "mcpg": {
      "command": "uvx",
      "args": [
        "mcpg"
      ],
      "env": {
        "MCPG_DATABASE_URL": "postgresql://user:pass@localhost:5432/mydb"
      }
    }
  }
}

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 MCPg?

MCPg is a production-grade MCP server for PostgreSQL that provides AI agents with over 100 tools for schema introspection, natural-language-to-SQL, query optimization, index tuning, performance analysis, multi-tenancy operations, and DBA tasks, prioritizing safety.

How to use MCPg?

Install via pip install mcpg or Docker. Configure connection settings and choose a transport: stdio for Claude Desktop, or HTTP/SSE/Streamable HTTP. Connect any MCP client to start using the tools.

Key features of MCPg

  • Safety-first design: read-only by default, AST validation, strict sanitization
  • Broad PostgreSQL support: pgvector, TimescaleDB, PostGIS, Apache AGE
  • Production ready: connection pooling, SET ROLE multi-tenancy, read-replica routing
  • Prometheus metrics and structured audit logging with redaction
  • Multiple transports: stdio, HTTP, SSE, Streamable HTTP
  • Easy install: pip or Docker with OIDC + static auth and rate limiting

Use cases of MCPg

  • AI-assisted schema exploration and database introspection
  • Natural language to SQL query generation and optimization
  • Automated index tuning and performance analysis
  • Multi-tenant database operations with role isolation
  • Production DBA tasks with safety guards

FAQ from MCPg

What tools does MCPg provide?

Over 100 tools covering schema introspection, natural-language-to-SQL, query optimization, index tuning, performance analysis, multi-tenancy operations, and DBA tasks.

How does MCPg ensure safety?

It is read-only by default, performs AST validation on all SQL, strictly sanitizes identifiers, and requires opt-in gates for DDL, shell commands, and LISTEN operations.

What transports are supported?

MCPg supports stdio (for Claude Desktop), HTTP, Server-Sent Events (SSE), and Streamable HTTP.

Is MCPg production ready?

Yes, it includes connection pooling, SET ROLE multi-tenancy, read-replica routing, Prometheus metrics, structured audit logging, OIDC + static authentication, and rate limiting.

What are the runtime requirements?

Python (psycopg3 driver), a PostgreSQL database, and optional extensions like pg_stat_statements for performance features.

评论

数据库 分类下的更多 MCP 服务器