Lirep: the quantitative repertoire builder

Most repertoire tools help you remember your moves. Lirep also tells you how good your repertoire is, in numbers, against the moves real players choose, and helps you decide what to prepare next.

Measure your repertoire

See the moves that matter

In the study editor, Stockfish's arrows show two things at once: the colour says how good a move is, compared with Stockfish's best move, and the thickness says how often it is played in the Explorer. Popular mistakes, the moves your opponents actually get wrong, stand out at a glance, and Stockfish also evaluates the popular moves its own top lines leave out.

The arrows: five colours for how good a move is, and thicker arrows for moves played more often best good inaccuracy mistake blunder rarely played sometimes most of the games
Colour: winning chances lost against Stockfish's best move. Thickness: the move's share of the games in the Explorer.

Remember it, share it, compare it

Based on Lichess

This website is based in part on source code and assets from Lichess, including its open-source chessboard library, Chessground. Lirep.org also uses Stockfish for chess analysis. It is an independent project and is not an official Lichess service.

Like the code of Lichess, the code of Lirep is released under the GNU AGPL-3.0-or-later.

No warranty

This software is provided without warranty, under the terms of its applicable licenses.

Created by

Lirep.org was built by Urbain Vaes with the assistance of AI agents.

Contact

Questions, bug reports and feedback are welcome at contact@lirep.org.

Similar tools

Lirep is one of several free and open-source tools for building and drilling an opening repertoire. Its unique features include:

For comparison, here is a short map of similar free software projects.

ToolLog inOpening explorerStockfishSpaced repetitionMetricsExtra features
Lichess studies (source)
Dynamic website · 0 €
Since Feb 2012 · ★ 18,807
AGPL-3.0-or-later
None
  • Annotations and comments on moves
  • Collaborative editing
OpeningTree (source)
Dynamic website · 0 €
Since Nov 2019 · ★ 476
GPL-3.0
  • Win/draw/loss percentages per move
  • Opening tree from any player's games (for scouting an opponent)
Listudy (source)
Dynamic website · 0 €
Since Jun 2020 · ★ 398
AGPL-3.0
  • Progress through each study
  • Achievements
  • Tactics, including blind and pieceless tactics
  • Endgames against Stockfish
  • Many public studies
En Croissant (source)
Desktop application · 0 €
Since Sep 2022 · ★ 1,926
GPL-3.0
  • Win/draw/loss percentages per move
  • Local game databases with position search
  • Full analysis toolkit
ChessRepeat (source)
Dynamic website · 0 €
Since May 2024 · ★ 8
No licence ⓘThe README calls the project open source, but the repository has no licence file, so its code isn't free software in the legal sense.
  • Memory scheduleA calendar of the reviews due over the next 20 weeks.
  • Moves due now
  • Real-time collaborative training
  • Comments on moves
  • PGN and JSON export
ChessStack (source)
Dynamic website · > 0 €Free with one repertoire, a small fee for more.
Since Feb 2026 · ★ 10
AGPL-3.0-or-later
  • Health scoreFrom 0 to 100: the average of coverage (100 minus 5 per gap against common opponent moves), share of learned positions, share of positions not currently due, and share without repeated errors.
  • Accuracy trendShare of correct answers in each of your last 14 drill sessions.
  • Trouble spotsThe 5 positions you have failed most often (at least 3 lapses).
  • StreakNumber of consecutive days with a completed drill session.
  • Opponent preparation from their games
  • Game review against your repertoire
  • Puzzles
Tabia (source)
Static website · 0 €
Since Jun 2026 · ★ 23
MIT
  • Correct and wrong movesCounted per drill session. A move played after asking for a hint doesn't count as correct.
  • Several drill modes
  • Notes on moves
Lirep (source)
Dynamic website · 0 €
Since Sep 2026
AGPL-3.0-or-later
  • Win probabilityExpected score (win % plus half the draw %) of your whole repertoire, assuming the opponent plays moves with the probabilities given by the explorer.
  • Expected evaluationAverage Stockfish evaluation at the end of your prep, weighted by how likely the opponent is to reach each line.
  • CoverageShare of games still following your lines after each opponent reply, move by move.
  • KnowledgePractice score per position that fades over time like real memory, so it shows what you can recall right now.
  • Sharing, with one-click import of others' openings

Gathered from each project's own pages and repositories in October 2026; features and prices change, so check them directly. Lirep is independent and not affiliated with any of these projects.

Want to know more?

The documentation explains how each part of Lirep works, down to how the expected score is calculated.