Feature
AI move explanations grounded in the engine
By Atulya GuptaMost engines tell you a move was a blunder but not why. The Audit Chess move explainer closes that gap: for a reviewed game it writes a short, plain-English comment on the moves that matter - what your move gave up, the threat you missed, the tactic that refuted it, and the stronger idea. Every sentence is grounded in a real Stockfish analysis of the position, so the explanation matches the evaluation rather than inventing a story. It detects hanging pieces with a static-exchange check, spots one-move threats, and names the refutation line, then writes the comment from those verified facts alone - no language model guesses, so it explains only what is actually on the board. It is free, needs no account, and the engine work runs in your browser.
Explanations that match the engine, not guesses
A language model on its own will happily narrate a plausible-sounding but wrong reason a move failed. Audit Chess avoids that by building a factual profile of the position first - the evaluation swing, whether a piece is actually hanging (checked with a static exchange evaluation, not a guess), the concrete threat, and the refutation the engine finds - and then writes the comment directly from those facts with fixed, tested rules. No language model is asked to guess, so the commentary cannot say anything the engine did not verify.
The result reads like a coach talking you through the game: "this drops the exchange to the knight fork on f5; Rc7 first held everything together" - tied to the eval that proves it, not a vibe.
What it explains
The explainer focuses on the moves that changed the game: inaccuracies, mistakes and blunders, the critical turning point, and the occasional brilliant or only-move. For each it names what was lost, the threat or tactic involved, and the better move, so you learn the idea rather than just the label.
It runs on top of the move-by-move game review, so the labels (best, mistake, blunder) and the written explanation come from the same analysis pass.
Related features
Frequently asked questions
How does Audit Chess explain why a move was a mistake?+
It runs the position through Stockfish, finds what your move gave up - a hanging piece, a missed threat, a refuting tactic - and writes a plain-English reason plus the better move, grounded in the engine eval so the explanation matches the evaluation.
Does the AI make up reasons for moves?+
No. The explainer builds a verified fact profile from a real Stockfish read - eval swing, hanging-piece check via static exchange evaluation, concrete threats and refutations - and the comment is written from those facts alone, with no language model guessing, so it explains what is actually on the board.
Is the move explainer free?+
Yes. It runs on top of the free game review, the engine work happens in your browser, and no account is required.