Manual Decoding Techniques

1. Every letter is swapped with its mirror opposite: A↔Z, B↔Y, C↔X, and so on down to M↔N.
2. Non-letters (spaces, punctuation) are left untouched.
3. The mirror is fixed and uses no key — applying the exact same swap a second time restores the original. Seal and unseal are the same action.

1. Each letter/number has its own fixed dot-dash pattern (reference table below).
2. Read one group at a time between spaces, decode it, then move to the next.
3. A / means start a new word.

1. Lay out the alphabet A–Z in a fixed 5×5 grid, sharing one cell between I and J (below).
2. Each two-digit code is (row, column) — first digit is the row, second is the column.
3. Look up the letter at that row/column and move to the next pair.

1. Guess a small number of rails, starting at 2.
2. Sketch that many rows and mark out a zigzag path across the message's full length (down to the bottom row, back up to the top, repeat).
3. Split the sealed text into chunks whose lengths match how many marks landed on each row.
4. Place each chunk back into its row, then read the whole thing off by following the zigzag path.
5. If it's still gibberish, try the next rail count — there are rarely more than a handful of reasonable options.

Each pair has two symbols. In this system, the letters A–F aren't real letters — they're just extra digits, standing in for the numbers 10 through 15 (there aren't enough plain digits to count that high with just one symbol, so letters fill the gap):

To read a pair: take the first symbol's value and multiply it by 16, then add the second symbol's value. That gives you one plain number between 0 and 255.

When this locker seals a message, it adds a chosen number (the "dial offset") to every one of these values. To undo it, do the opposite: pick a small number to try, and count backward by that amount from each value you worked out in Step 1. If counting back takes you past 0, wrap around by adding 256 back on.

Every capital letter has a fixed number, starting at A=65 and counting up in order:

Find your Step 2 result in this chart to get the letter. If none of the results land near this 65–90 range at all, the offset you guessed is probably wrong — try a different number and start over from Step 2.

1. If you know or suspect the keyword: write it out, repeating, under the sealed message — one keyword letter per plaintext letter, skipping non-letters.
2. For each pair, count backward through the alphabet by the keyword letter's position (A=0, B=1, C=2...) to undo the shift.
3. If you don't know the keyword: look for any repeated group of letters in the sealed text. The distance between repeats is often a multiple of the keyword's length — a strong hint at how long it is. From there, test short, likely words.

1. Build a 5×5 grid from the keyword: write the keyword's letters first with no repeats, then fill the rest of the alphabet in order (I and J share one cell).
2. Take the sealed text two letters at a time and find each letter's row and column in the grid.
3. Same row → replace each with the letter to its immediate left (wrapping around the row).
4. Same column → replace each with the letter immediately above it (wrapping around the column).
5. Otherwise → swap each letter with the one in its own row, but in the other letter's column (forming a rectangle).