Opening Inaccuracies Tool

When I started playing chess in November 2020, I progressed quickly and got hooked almost immediately. I bought courses, watched far too much YouTube, bought books (and even read some of them), and tried to absorb as much structured knowledge as I could. The goal was simple: reach a solid enough level that I could one day teach my daughter properly and explain not just what to play, but why.
By 2024, I had returned to programming after a long break and wanted to combine the two interests. I was not trying to build anything flashy or commercial. I simply wanted tools that would help me understand my own chess in a more honest way, especially the parts of my game that felt familiar but were quietly holding me back.
The first idea I wrote down was something I later called Opening Inaccuracies, and it came from a very specific experience. I was challenged by a stronger friend about a line in the Caro-Kann that I kept playing. It was not really a blunder, but it was also not particularly good, and yet I kept playing it anyway. That moment made me wonder where else I might be repeating slightly bad opening choices out of habit rather than understanding, and whether it was possible to identify those patterns across all my games, not just in one line or one opening, but everywhere I play.
In theory, the idea was straightforward. Import my games from Chess.com and Lichess, run them through Stockfish, highlight the opening moves where I consistently underperform, and then show me stronger alternatives that I could actually explore and learn from.
In practice, I was not ready to build that yet.
At the time, I had only just returned to programming, and trying to handle game imports, large-scale engine analysis, and the general complexity of building chess software all at once was simply too much. Rather than forcing it and ending up with something half-finished, I stepped back and spent time building smaller chess tools instead. Over time, that turned into more than 50 small applications, experiments, and prototypes, each one teaching me something about engines, data, performance, and how players actually interact with analysis.
Eventually, it felt like the right time to return to the original idea and do it properly.
That is how Opening Inaccuracies finally came together.

What the Tool Does

The current version allows you to:

• Import your games from Chess.com or Lichess
• Configure and run Stockfish analysis over the opening phase
• Identify sub-optimal opening moves you play in real games
• Explore stronger engine-recommended alternatives, in context rather than in isolation

The focus is not on one-off mistakes. It is on patterns: the moves you keep choosing that quietly make your position worse before the middlegame even begins.
This is still a first draft of the tool. The full version is planned for the end of January, with deeper analysis, better filtering, and more ways to learn from the results.
If you want to try it out, you can find it here:
https://test.chessboardmagic.com/openinginaccuracies
I would love to hear what you think, especially if this highlights opening habits you recognise from your own games.

Kind regards,
Toan (@HollowLeaf)