Every detective in the world got these same crime scenes on 2026-11-08 — each difficulty published twice, once untimed and once as time attack. Each case is procedurally generated and then proved to have exactly one logical solution before it is published, including the cases where somebody is lying. You place every suspect back on the map using only their testimony: exactly one person per row, never two in a column. Whoever is left alone with the victim did it.
A 5×5 grid across 3 rooms: Seed bank, Orchid room, Potting shed. Six statements, five squares nobody could stand on, and no assumption anywhere: elimination alone closes it.
Cracked by 2 detectives so far.
A 5×5 grid across 4 rooms: Orchid room, Seed bank, Potting shed, Shed. Seven statements, seven squares nobody could stand on, and enough testimony that the surplus is what pins the last two down.
A 6×6 grid across 5 rooms: Orchid room, Palm house, Arbor, Potting shed, Shed. Eight statements, nine squares nobody could stand on, and no assumption anywhere: elimination alone closes it.
Cracked by one detective so far.
A 6×6 grid across 4 rooms: Orchid room, Apiary, Fountain, Arbor. Seven statements, eight squares nobody could stand on, and no assumption anywhere: elimination alone closes it.
An 8×7 grid across 6 rooms: Apiary, Palm house, Seed bank, Terrace, Fountain, Arbor. Ten statements, twelve squares nobody could stand on, almost as many rooms as suspects, so a room claim nearly names a square, and enough testimony that the surplus is what pins the last two down.
A 7×7 grid across 5 rooms: Arbor, Seed bank, Orchid room, Apiary, Terrace. Ten statements, nine squares nobody could stand on, and one round of assume-and-test after elimination ran dry.
Cracked by one detective so far.
A 10×8 grid across 7 rooms: Shed, Apiary, Potting shed, Seed bank, Fountain, Palm house, Terrace. Twelve statements, sixteen squares nobody could stand on, almost as many rooms as suspects, so a room claim nearly names a square, and one round of assume-and-test after elimination ran dry.
A 9×8 grid across 7 rooms: Seed bank, Terrace, Shed, Palm house, Apiary, Potting shed, Arbor. Twelve statements, sixteen squares nobody could stand on, almost as many rooms as suspects, so a room claim nearly names a square, and one round of assume-and-test after elimination ran dry.
A 9×9 grid across 7 rooms: Apiary, Fountain, Orchid room, Terrace, Palm house, Seed bank, Arbor. Seventeen statements, sixteen squares nobody could stand on, and two rounds of assume-and-test after elimination ran dry — and one of the nine was lying throughout.
The twist: one suspect is an accomplice — every statement they make is false; the accusation also demands the motive, not just the name.
A 10×9 grid across 7 rooms: Apiary, Palm house, Potting shed, Seed bank, Shed, Terrace, Fountain. Thirteen statements, nineteen squares nobody could stand on, and three rounds of assume-and-test after elimination ran dry.
The grid size is the size of the crime scene, so a 10×9 expert case is a far larger search than a 5×5 beginner one — but the difficulty is really carried by the clues, not the square count. A beginner case tells you almost directly where somebody stood. An expert case tells you where somebody was not, three times, and expects you to do the rest — and on half of them one of those statements is a lie told to protect the killer. If a case lists a motive, naming the right person is not enough on its own; you have to prove why.
How to play · Solving cases with a liar · Techniques
Every day's cases — the calendar · ← the cases for 7 November