Baconian cipher encoder and decoder

An incomplete final group, an unknown five-symbol pattern, or a character that is not a, b, or a separator fails with invalid_input instead of guessing. Use it free in the browser for history-of-crypto lessons, steganography demos that still need the bare alphabet mapping, and reversible puzzle worksheets, or call the API at $0.002 per successful request when pipelines need reproducible Baconian transforms.

What the Baconian cipher is and what this tool returns

The Baconian cipher, also called Bacon’s cipher, is a classical biliteral substitution in which every letter of a message is replaced by a five-symbol word drawn from a two-letter alphabet usually written a and b. The modern twenty-six letter table treats those symbols as binary digits with a equal to zero and b equal to one, so A is aaaaa, B is aaaab, and so on through Z as bbaab. Historically the codes were meant to be hidden inside ordinary text by changing typeface or capitalization; this product only exposes the bare encode and decode table so teachers, puzzle authors, and test authors can work with the codes themselves without simulating fonts. That fixed mapping is exactly what this endpoint applies: there is no keyword stream, no modular shift, and no optional alphabet file that could drift between locales. The JSON response always includes result for the transformed string, encoding fixed to bacon, mode set to encode or decode, input_length and output_length for the raw strings you sent and received, letter_count for how many A–Z letters were produced or consumed, and a formula string that restates the binary table rule for documentation and automated checks. There is no streaming path and no silent fallback to Morse, Base64, or a Vigenère keyword: either the payload is valid for the chosen mode, or the request fails with invalid_input. If you need polyalphabetic ciphers, ROT variants, or modern authenticated encryption, use those dedicated endpoints; this landing answers only the five-symbol Baconian A/B contract that crypto history chapters already describe.

How text and mode are read and how incomplete groups fail

In decode mode every a or b is collected into a continuous stream while spaces, tabs, newlines, hyphens, underscores, commas, dots, slashes, and pipes are treated as harmless separators between groups. Any other character is rejected immediately. Each complete five-symbol slice must also exist in the twenty-six letter table; patterns outside A–Z codes fail as unknown groups. The preferred payload fields are text for the message or ciphertext and optional mode for encode or decode. Text accepts aliases such as input, value, s, message, plaintext, and ciphertext so forms and SDK clients can reuse existing column names. Mode defaults to encode when omitted and also accepts short aliases such as enc, encrypt, e for encode and dec, decrypt, d for decode; any other string fails validation. In encode mode only alphabetic letters are kept and mapped, case is ignored on the way in, and the result is lowercase a/b groups separated by single spaces; if no letter remains after filtering, the engine rejects the request with invalid_input so a digit-only string does not look like a successful empty cipher. After cleaning, the stream length must be a multiple of five: a leftover of one through four symbols means a group that is not five symbols, and that condition always raises invalid_input with an explicit incomplete-group message rather than padding zeros or inventing a letter. Missing text, empty text, non-string types, and unknown modes likewise raise invalid_input. Pair the free browser run on this page with the priced API when you need the identical mathematical result inside an automated pipeline. Invalid inputs return invalid_input and are not charged, which keeps validation cheap while successful transforms remain a fixed item price.

Where Baconian fits in lessons, puzzles, and test fixtures

On this page the work runs entirely in your browser; through the API each successful lookup costs $0.002, failed validation is not charged, and nothing is stored after the response returns. That split — free interactive exploration and a priced, identical API — matches the rest of the KIT tier-A catalog and lets you move from a classroom demo to an automated pipeline without changing the transformed result.