MySQL (read-only)

Lets Claude run safe, look-only SQL queries against a MySQL database — explore tables and pull data for analysis, with all edits/deletes blocked by design. Relevant only where MM data lives in a MySQL database you connect.

Claude Code only

How to use it

Once a connection is configured (in a credentials file you control, locked to your machine), ask a data question in plain English and Claude writes the SELECT and returns results — it will refuse INSERT/UPDATE/DELETE/DROP. Claude Code only: it opens a real database connection from your machine. You set up the credentials file; never paste DB passwords into chat.

How to invoke / keep it on

Ask to query / explore a MySQL database in Claude Code and it loads. It auto-picks the right database from your configured connection descriptions.

20 powerful examples for Melinda Maria

Copy, tweak, make them yours. Just describe the task — Claude loads the skill automatically.

01"Explore the schema of our connected MySQL database and tell me which tables hold order and customer data."

02"Pull the top 20 best-selling SKUs in the last 90 days from the MySQL database, read-only, and format it as a table."

03"Write a SELECT that joins orders to customers so I can see repeat-purchase rate — and confirm it can’t change any data."

04"Sanity-check a number from a connector against the source MySQL database with a read-only query."

05"Show me total revenue by month for the last year from the MySQL database, formatted for a quick chart."

06"List the tables and row counts so I understand how big each part of the database is before I query it."

07"Write a read-only query to find customers who bought a lariat but never a charm, for a cross-sell idea."

08"Pull average order value by month and flag the highest and lowest months."

09"Find the most-returned products from the MySQL data so ops can investigate, read-only."

10"Write a query for new vs returning customer revenue split, and explain how you defined “returning.”"

11"Show me the schema for the orders table column by column so I know what’s available to analyze."

12"Pull a list of the top 50 customers by lifetime spend, names redacted to IDs, as a table."

13"Write a safe query for units sold per SKU last quarter and confirm it’s a pure read."

14"Find any obvious data-quality issues (nulls, weird dates) in the orders table with read-only checks."

15"Compare this month’s order count to the same month last year straight from the database."

16"Write a query to see which states or regions drive the most revenue."

17"Pull repeat-purchase interval (days between first and second order) for a cohort, read-only."

18"Show me discount usage: how much revenue came from discounted orders versus full price."

19"Help me export a read-only query result to CSV so I can share it with the team."

20"Double-check a Shopify-reported figure against the MySQL source with a SELECT, and tell me if they match."

← Back to Databases