Daily puzzle, 7 November 2026

Step 1 was public. Each step's solution, SHA-256'd, is the AES-256-GCM key for the next step. The full puzzle with its encrypted steps is at /api/puzzles/2026-11-07.txt.

STEP 1 OF 5 — INDUCTION

Each grid is rows of digits; 0 is background, 1-9 are colours. Every output below was made from its
input by the same hidden transformation. Work out the transformation and apply it to the test input.

Example 1 input (15 rows x 14 cols):
00000000666000
00000000000000
00033300004444
00030300404444
00033300000000
44030307000000
44033300000700
00000000000000
00000000000007
00000000000000
00006660000000
00006060000000
00006660040000
00006060000000
00006660000000

Example 1 output (15 rows x 14 cols):
44400000000000
00444440000000
00455540222200
00450542222200
00455540000000
44450547000000
44455547000000
00444440000000
70000000000000
00000000000000
55500000000000
50500000000000
55540000000000
50500000000000
55500000000000

Example 2 input (14 rows x 15 cols):
000000000000000
006660000000000
606060000700000
006660000000000
006060006000000
006660000000000
000000000000444
066033333000444
066030303000000
066033333000000
000000000000003
003000000000000
033300066600000
030000006000000

Example 2 output (14 rows x 15 cols):
000000000000000
055500004440000
550500004740000
055500004440000
050500004000000
055500044440000
000000042220000
665555542220000
665050544440000
665555500000000
300000000000000
030000000000000
333222000000000
300020000000000

Example 3 input (18 rows x 19 cols):
0000000000000000000
4444444003000060000
4000004000000000000
4000004000033333000
4003004000030003000
4000004000030003006
4000004003033333000
4444444000000000000
0000000000000000000
0000000004444406007
0077770004040400000
0077770004444400060
0077770000000000000
0000000000066600044
0000000000000600040
0000000700000000040
0000000000000330000
0003000000000330000

Example 3 output (18 rows x 19 cols):
0000000000000000000
4444444360000000000
4000004000000000000
4000004044444000000
4300004040004000000
4000004040004600000
4000004444444000000
4444444000000000000
0000000000000000000
0000555556700000000
5555505050000000000
5555555556000000000
5555000000000000000
5554400000000000000
0054000000000000000
7004000000000000000
0220000000000000000
2220000000000000000

Test input (16 rows x 15 cols):
666000000000777
000000000000007
000000300000007
000000000000000
000000000040300
033300000000000
033300333330000
000000300030000
000000333330000
600000000000000
000066006666600
000066006060600
000066006666600
000000000000000
000000000030000
000000000000000

Solution: the test output as a Canonical grid string: each row as its digits, rows joined by "/", no spaces (e.g. 0120/0300/0000).

Step 2 is encrypted with AES-256-GCM. Key = SHA-256 of this step's solution string (UTF-8).
The blob is base64 of: 12-byte IV || ciphertext || 16-byte auth tag. If the tag verifies, your solution is right.

Final answer

3debab33354bc73061bf65aa392f8f20e19f4c2891aaf5e270cfad96f8a6725d

Step 1 solution

222000000004777/000000000004007/300000000004007/000000000444444/000000000444300/444000000444000/444444440000000/0004000…

Step 2 solution

45d5238579e0635c48ea26e8

Step 3 solution

0000000000000000000/0000000000000005000/0000005555500000002/0000025050588805550/0000005555500005050/8880000000000005550…

Step 4 solution

15197325352063880

Step 1 rule

frame(unique-color->6) → gravity(left,all) → recolor(wide->2) → recolor-holes(1->4,2+->5)

Step 3 rule

recolor(has-hole->5) → gravity(right,marker-color) → rays(color-8,left)

VM kernels

#{(i,j): 1<=j<=i<=n, gcd(i,j)=1} = sum phi(i), n=4165737875+(prev&65535) → sum_{i<4237816594102} floor((522102i+b)/934728) → sum_{i<=n} floor(n/i), n=448415448330229+(prev&65535)

Step 2 world

shortest win: 54 moves, sliding moves

Fairness: the commitment 0e323cbab66a76c34fa75972f8a420f2a539d4bdd83538d65f9fd8119110c071 was published before release; check sha256("hunt|" + id + "|" + answer + "|" + salt), salt revealed when the puzzle closes with salt 34459b439f8b3b8d63756980be9139dc.

Practise on it: POST https://logicpot.com/api/puzzles/2026-11-07/practice with {"answer": "..."}.