Simon Willison built a browser-based SQLite query explainer after Julia Evans admitted she still cannot read a query plan. The tool runs SQLite in Python via Pyodide compiled to WebAssembly, then wraps both EXPLAIN and EXPLAIN QUERY PLAN output in plain-language annotations. No server. No install. It runs entirely client-side.
The build itself is worth noting: Willison used Fable, an AI coding tool, to generate it via a single pull request. The technical stack, SQLite inside Python inside Pyodide inside WebAssembly inside a browser, is a real demonstration of how far the Pyodide runtime has come for interactive tooling.
Willison flags the caveat directly: he does not know enough about SQLite internals to verify the AI-generated explanations are accurate. That honesty is the reason to read the original and try the tool yourself. If you work with SQLite query performance, this is either useful or a useful test of where LLM-generated explanations break down.
[READ ORIGINAL →]