Free beta · Postgres & MySQL

Find slow SQL before it ships

Paste a query and get plain-English performance findings, a health score, and actionable fixes in seconds — no database connection required.

analysis · postgres

SELECT * FROM users
WHERE email LIKE '%@gmail.com'
ORDER BY created_at;

Score 26· 4 findings · lower is better (0 = clean)

high
LEADING_WILDCARD_LIKE
Pattern “%@gmail.com” can’t use a B-tree index — forces a full scan.
medium
SELECT_STAR
Project only the columns you need to cut network/IO cost.
low
UNINDEXED_ORDER_BY
Sorting every row is expensive — add LIMIT or a covering index.

✦ AIFor substring search in Postgres, create a GIN index with gin_trgm_ops, or use full-text search.

Example output — your results will match your query.

How it works

Drop in any Postgres or MySQL query. Nothing connects to your database — it analyzes the query text.

29 rules flag performance anti-patterns, each with severity and a health score from 0–100.

Every finding includes an actionable recommendation — and an AI explanation of why it matters.

SELECT *, leading-wildcard LIKE, deep OFFSET, cartesian joins, NULL-unsafe NOT IN, functions in WHERE, and more.

Dialect-aware parsing — same UI, both engines, including dialect-specific syntax.

Plain-English “why it matters” for each fix — grounded in the deterministic finding, never invented.

Paste EXPLAIN (FORMAT JSON) and get plan-level insights and index recommendations.

Export any analysis as a PDF and revisit your past runs anytime.

Queries are encrypted at rest, PII is masked, and we never store database credentials.

This is a beta — I’d genuinely love your feedback, bug reports, or any question about how it works. Reach me directly:

Or click the ✦ Ask button for instant answers about the product.

It’s free during beta. Paste a query and see what comes back.