# planetscale/mcp [Official] [Health: Active]

**Category:** 🗄️ Databases  
**Repository:** https://github.com/planetscale/cli?tab=readme-ov-file#mcp-server-integration  
**GitHub Stars:** 672  
**Views:** 2  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/planetscale-mcp

## Description
The PlanetScale CLI includes an MCP server that provides AI tools direct access to your PlanetScale databases.

## Claude Desktop Quick Installation
No generated config for this server: it installs through its own setup steps.

The project's own setup steps (from https://github.com/planetscale/cli?tab=readme-ov-file#mcp-server-integration), copied verbatim. Run them in order; they are not an mcpServers config.

1. Install:

```sh
brew install planetscale/tap/pscale
```

## Documentation

## What planetscale/mcp does

planetscale/mcp connects AI agents to PlanetScale databases through the Model Context Protocol. It is provided as part of the PlanetScale CLI ecosystem and is intended to expose PlanetScale database access as agent-usable tools.

The integration is relevant for teams using PlanetScale as their database platform and wanting an AI client to interact with those databases through MCP rather than through a custom application integration.

## How it works

PlanetScale provides a hosted MCP server for this integration. An MCP-compatible AI client connects to the hosted PlanetScale endpoint, after which the agent can use the server's database-access tools. The repository points to PlanetScale's MCP documentation for the connection details and usage instructions.

This server is distinct from the broader `pscale` command-line interface, which also handles PlanetScale concepts such as branches and deploy requests. The MCP integration focuses on making PlanetScale database access available to AI agents.

## Setup and configuration

Use the hosted PlanetScale MCP server at:

`https://planetscale.com/docs/connect/mcp`

The PlanetScale CLI itself can be installed on macOS with Homebrew, on Windows with Scoop, or on Linux and other platforms using release binaries. The CLI also has Docker images. These installation paths apply to `pscale`; connecting an AI client to the hosted MCP server follows PlanetScale's MCP setup instructions.

For local CLI usage, some commands require a MySQL 8 client in the system path. That requirement belongs to the CLI and is separate from the hosted MCP connection.

## Tools and capabilities

- Provides AI tools with direct access to PlanetScale databases.
- Exposes PlanetScale database operations through MCP.
- Offers a hosted endpoint for MCP clients.
- Fits database workflows built around the PlanetScale platform.

## Limitations and notes

The PlanetScale CLI includes additional functionality for branches, deploy requests, and other PlanetScale concepts; those CLI capabilities should not be assumed to be MCP tools.

The repository also links PlanetScale skills for operational workflows such as inventory, safety review, and schema recommendations. Those skills are a separate installation from the MCP server and should be evaluated independently.

_Full upstream README: https://allmcps.com/mcp/planetscale-mcp/readme_