SQL and NoSQL Database MCP Servers

Your AI just wrote a beautiful ORM query — against a table that does not exist. It assumed users.email is unique when your schema allows duplicates. It generated a migration that drops a column still referenced by three views. Every one of these failures happens because the AI is guessing at your database schema instead of reading it.

Database MCP servers solve this by giving your AI direct access to schema metadata and the ability to run read-only queries. The AI stops guessing and starts working with your actual data structure.

What You’ll Walk Away With

• Setup instructions for PostgreSQL, MongoDB, SQLite, and Supabase MCP servers
• A clear decision framework for choosing the right database MCP for your stack
• Security patterns for read-only access and query sandboxing
• Prompts that demonstrate schema-aware code generation

PostgreSQL: Prisma MCP and Supabase MCP

For TypeScript teams, the Prisma MCP server is the most natural fit. It integrates with your existing Prisma schema and allows the AI to query data, inspect schemas, and even manage migrations.

Cursor

{
  "mcpServers": {
    "prisma": {
      "command": "npx",
      "args": ["-y", "prisma", "mcp"]
    }
  }
}

Claude Code

{
  "mcpServers": {
    "prisma": {
      "command": "npx",
      "args": ["-y", "prisma", "mcp"]
    }
  }
}

Codex

[mcp.prisma]
transport = "stdio"
command = "npx"
args = ["-y", "prisma", "mcp"]

For teams using Supabase, the Supabase MCP server is aware of Row Level Security policies and the full Supabase ecosystem including auth, storage, and edge functions.

Tip

Copy-paste prompt for schema-aware development:

“Inspect the database schema. List all tables, their columns, types, and relationships. Then write a Drizzle ORM query that fetches all users who signed up in the last 30 days along with their most recent order, including the order total and status.”

MongoDB MCP

For document databases, the MongoDB MCP server provides schema inspection and JSON querying. It helps the AI navigate semi-structured collections and understand embedded document patterns.

Cursor

{
  "mcpServers": {
    "mongodb": {
      "command": "npx",
      "args": ["-y", "mongodb-mcp-server"],
      "env": {
        "MONGODB_URI": "mongodb://localhost:27017/mydb"
      }
    }
  }
}

Claude Code

{
  "mcpServers": {
    "mongodb": {
      "command": "npx",
      "args": ["-y", "mongodb-mcp-server"],
      "env": {
        "MONGODB_URI": "mongodb://localhost:27017/mydb"
      }
    }
  }
}

Codex

[mcp.mongodb]
transport = "stdio"
command = "npx"
args = ["-y", "mongodb-mcp-server"]

[mcp.mongodb.env]
MONGODB_URI = "mongodb://localhost:27017/mydb"

SQLite MCP

SQLite is the reliable choice for local-first development and prototyping. Let the AI experiment with schemas and build internal tools without touching production data.

Cursor

{
  "mcpServers": {
    "sqlite": {
      "command": "uvx",
      "args": ["mcp-server-sqlite", "--db-path", "./dev.db"]
    }
  }
}

Claude Code

{
  "mcpServers": {
    "sqlite": {
      "command": "uvx",
      "args": ["mcp-server-sqlite", "--db-path", "./dev.db"]
    }
  }
}

Codex

[mcp.sqlite]
transport = "stdio"
command = "uvx"
args = ["mcp-server-sqlite", "--db-path", "./dev.db"]

Tip

Copy-paste prompt for migration generation:

“Read the current database schema and compare it to the TypeScript types defined in src/lib/db/schema.ts. Identify any mismatches — missing columns, type differences, or missing indexes. Generate a SQL migration script that brings the database in line with the TypeScript schema.”

Choosing the Right Database MCP

Stack	Recommended MCP	Why
TypeScript + Prisma	Prisma MCP	Native integration with your ORM and migration system
Supabase	Supabase MCP	Respects RLS policies, includes auth and storage tools
MongoDB	MongoDB MCP	Schema inference, JSON querying, collection analysis
SQLite / Local dev	SQLite MCP	Zero-config, perfect for prototyping
Multiple databases	dbhub	Universal connector supporting PostgreSQL, MySQL, SQLite, and more

Caution

Always start with read-only access. Database MCP servers that can run arbitrary queries can also run DROP TABLE. Scope the database user to SELECT-only permissions until you have established trust in the workflow. Use a development or staging database, never production, for initial experimentation.

Tip

Copy-paste prompt for data analysis:

“Query the orders table for the last 90 days. Group by product_id and calculate the total revenue, average order value, and order count for each product. Sort by revenue descending and present the top 10 products in a formatted table.”

When This Breaks

Connection refused. Make sure your database is running and accessible from localhost. Check that the connection string includes the correct port, database name, and credentials.

AI writes destructive queries. If the AI runs UPDATE or DELETE when you only expected SELECT, the database user has too many permissions. Create a read-only database role for MCP access.

Schema inspection returns stale data. Some MCP servers cache schema metadata. If you run a migration, restart the MCP server to refresh the schema cache.

Slow queries time out. Complex analytical queries on large tables may exceed the MCP server’s timeout. Add appropriate indexes or break the query into smaller pieces.

What’s Next

Documentation Context Keep your AI grounded in current library documentation.

Version Control MCP GitHub and GitLab integration for AI-assisted development.