KiloClawPowered by OpenClaw
Back to Cookbook

Slow SQL Query Diagnoser and Optimizer

aka “Query Surgeon”

Diagnose and fix slow SQL without being a DBA

Paste a slow query and get back a diagnosis in plain English — what's slow, why it's slow, and which fixes are most likely to help. Reads EXPLAIN plans so you don't have to. Suggests indexes, rewrites, and pre-aggregation strategies.

CommunitySubmitted by CommunityWork1 min
Try in KiloClawFree 7-day trial

PROMPT

Create a skill called "Query Surgeon". When I paste a SQL query that's running slowly, analyze it for performance issues. If I include EXPLAIN ANALYZE output, parse it and translate every node into plain English (e.g., "Sequential Scan on orders — scanning all 2.3 million rows because there's no index on customer_id"). If I don't include EXPLAIN output, infer likely bottlenecks from the query structure. For each issue found, explain: what's happening, why it's slow, and how to fix it. Generate an optimized version of the query with comments. Suggest CREATE INDEX statements where applicable. If the query runs repeatedly, suggest a materialized view or pre-aggregation table. Estimate the likely improvement range for each suggestion and state the assumptions behind it.

How It Works

Your query takes 45 minutes and you have no idea why. EXPLAIN ANALYZE

produces a wall of text that only DBAs understand. This skill translates

query plans into plain English, identifies the specific bottleneck, and

generates an optimized version with likely impact ranges and clear assumptions.

What You Get

  • Plain-English diagnosis of why the query is slow
  • Identification of the specific bottleneck (full table scans, bad join order, missing indexes, correlated subqueries)
  • An optimized rewrite with inline comments explaining each change
  • Index recommendations with CREATE INDEX statements
  • Expected impact range for each suggestion, with assumptions called out
  • Pre-aggregation table suggestions for queries that run repeatedly

Setup Steps

  1. Ask your Claw to create a "Query Surgeon" skill with the prompt below
  2. Paste your slow query (and EXPLAIN output if you have it)
  3. Tell it which database engine you're using
  4. Get back a diagnosis and optimized query

Tips

  • Include the EXPLAIN ANALYZE output if you can — it makes the diagnosis much more precise
  • For queries that run on a schedule, the pre-aggregation suggestions save the most time
  • The skill will ask about table sizes and indexes if it needs more context
  • Works on any SQL dialect — Postgres, BigQuery, Snowflake, MySQL, Redshift, SQL Server
Tags:#sql#performance#optimization#databases

Related Recipes

Dialect Flip

Translate SQL between any two databases instantly

Paste SQL written for one database and get the equivalent for another. Handles the syntax nightmares — DATEADD vs DATE_ADD, QUALIFY, LIMIT vs TOP, string functions, NULL handling — across Postgres, MySQL, BigQuery, Snowflake, Redshift, SQL Server, and more.

Work1 min

Pandas Pit Stop

Find and fix the slow, memory-hungry parts of your pandas code

Paste your pandas script and get a performance audit — which operations are slow, what's eating memory, where you're using apply() when vectorized ops would be much faster, and where you should switch to Polars or DuckDB.

Work1 min

Proxy Factory

Auto-generate proxies the moment footage lands

Choppy playback, laggy scrubbing, and unusable timelines — even on strong machines — often comes down to editing long-GOP or heavy codecs at high resolution without proxies. This recipe generates proxies automatically whenever new camera originals appear, so every project starts edit-ready instead of debug-ready.

Creative20 min setup

Cache Janitor

Keep media caches from eating your SSD and slowing your apps

Media cache and disk caches balloon over time, causing storage emergencies and performance regressions. This recipe turns cache cleanup from a panic action into a controlled, scheduled maintenance job — for Adobe Premiere Pro, After Effects, and similar tools.

Creative10 min setup