MCP/PlanetScale MCP Server

PlanetScale MCP Server

Hosted Model Context Protocol server for PlanetScale databases, schema branching, query insights, and infrastructure metrics.

rag-data•Databases & Storage

GitHub120

License:Apache-2.0

VERIFIED

@PlanetScale

About this Server

The official PlanetScale Model Context Protocol (MCP) server connects AI agents directly to PlanetScale cloud databases and Vitess/Postgres infrastructure. It enables natural language schema exploration, read/write query execution with replica routing, schema change recommendations, error fingerprint tracking, query tag summaries, and deep Insights performance metrics.

Real-time Interaction

Direct low-latency RPC protocol communication with structured JSON schemas.

Scoped Sandboxing

Restricted capability bounds to prevent accidental modifications or leakage.

Security & Best Practices

OAuth & Service Tokens: Authenticate via browser OAuth or restricted service tokens for headless CI environments.

Replica Read Offloading: Default reads to replica instances to safeguard primary database performance.

Branch Isolation: Test migrations and experimental DDL on dedicated development branches prior to production deploy requests.

When to Use (and When Not To)

Best For

•Inspecting schemas, branches, and running SQL queries on PlanetScale databases
•Diagnosing database latency and slow query bottlenecks using PlanetScale Insights
•Reviewing schema index recommendations and error patterns without leaving the editor

Not For / Limitations

•Vector similarity search over unstructured document embeddings (use Chroma MCP instead)
•Non-relational document stores and Atlas streams (use MongoDB MCP instead)
•Local embedded in-memory Parquet analytics (use DuckDB / MotherDuck MCP instead)

Install & Setup

One-line terminal command

terminal command

$claude mcp add planetscale --transport http https://mcp.pscale.dev/mcp/planetscale

Exposed Tools & Endpoints

10 tools

Functions your AI agent can invoke via this Model Context Protocol server

planetscale_execute_read_query• callable tool

Executes read-only SQL queries (SELECT, SHOW, DESCRIBE, EXPLAIN) with optional replica or primary routing.

planetscale_execute_write_query• callable tool

Executes write SQL queries (INSERT, UPDATE, DELETE, or DDL) against a PlanetScale database branch.

planetscale_get_branch_schema• callable tool

Retrieves the complete schema and table definitions for a specific database branch.

planetscale_get_insights• callable tool

Fetches query performance insights, latency metrics, and row count trends for a database branch.

planetscale_list_databases• callable tool

Lists all databases belonging to a specified PlanetScale organization.

planetscale_list_branches• callable tool

Lists all development and production branches within a database.

planetscale_list_schema_recommendations• callable tool

Lists schema suggestions including missing indexes, redundant indexes, and primary key health.

planetscale_list_query_error_patterns• callable tool

Lists aggregated failing query patterns, error messages, and occurrence frequencies.

planetscale_get_branch_metrics• callable tool

Fetches infrastructure telemetry: CPU, memory (RSS/cache), disk, replication lag, and connections.

planetscale_search_documentation• callable tool

Searches official PlanetScale documentation and technical guides.