DEV Community

curatedmcp for CuratedMCP

Posted on • Originally published at curatedmcp.com

PlanetScale MCP: Deploy Schema Changes Without Downtime via AI

Install guide and config at curatedmcp.com

PlanetScale MCP: Deploy Schema Changes Without Downtime via AI

PlanetScale MCP connects your AI agents directly to PlanetScale's serverless MySQL platform, unlocking a workflow that's been impossible with traditional databases: propose schema changes, test them in isolated branches, and deploy to production with zero downtime.

Whether you're using Claude, Cursor, or Windsurf, this server gives your AI agent full control over database operations—from executing queries to managing the branching and deploy request system that makes PlanetScale unique. It's the database layer your AI agent was missing.

What It Does

PlanetScale MCP exposes core database operations as tools your AI can invoke:

  • Execute queries across any of your PlanetScale databases
  • Create and manage branches for safe, isolated schema changes
  • Open deploy requests to safely migrate schema changes to production without blocking reads or writes
  • Monitor performance with query insights and slow query analysis
  • Manage credentials securely—connection strings, passwords, and access controls
  • Inspect deployments with rollback capabilities if something goes wrong
  • Analyze cache performance on Boost-enabled databases
  • Handle data migrations and imports

The killer feature: your AI can now propose a schema migration, create a branch to test it, validate the change works, and open a deploy request—all without you writing a single migration script or worrying about downtime. It's VCS for your database.

How to Install

npx -y @planetscale/mcp
Enter fullscreen mode Exit fullscreen mode

Add to your Claude Desktop config:

{
  "mcpServers": {
    "planetscale-mcp": {
      "command": "npx -y @planetscale/mcp"
    }
  }
}
Enter fullscreen mode Exit fullscreen mode

Authenticate with your PlanetScale service token. Full setup instructions are in the PlanetScale MCP docs.

Real-World Use Cases

  • Rapid feature development: Ask Claude to add a new column to your schema, test it in a branch, and automatically open a deploy request—no manual SQL scripts required.
  • Performance optimization: Have your AI analyze slow queries from Insights, propose index changes, test them in isolation, and deploy safe improvements.
  • Data migrations at scale: Let your agent handle imports and migrations across branches while your production database stays live and serving traffic.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Top comments (0)