Everything below this card is a live channel. The guided lab moves one switch at a time; the crate re-runs a real two-party handshake and two real encrypted messages, and every panel underneath — the verdict, the adversary matrix, the bytes on the wire — is the result of that run. Nothing here is a mock-up, so if you change a switch yourself the numbers change with it. Predict before you apply. That is the step the lab is built around.
Guided lab
Reveal the answer, once you have written yours down
Transmission
Layers
The adversary
What the adversary recovers
Which adversary
A1–A5, computed for this configuration“Channel holding” is never universal. It holds against a named adversary, for a named asset — A4 is a adversary and needs no access today. Definitions are in THREAT_MODEL.md.
What leaks regardless
true in every configuration aboveis what stays true about a message when its contents do not. No switch on this page moves any of it.
On the wire
The intended recipient
Handshake
Both sides hash every handshake message into a , which is signed and then bound into every frame as . Verifying that signature against a key known in advance is .
Inspect internals — transcript hash, root key, per-frame message keys
Terms
one sentence eachDotted terms elsewhere on the page open the same definition on hover or keyboard focus.
Why this stack, and whose it was
The Navajo code was a Type One Code: Navajo words stood for English letters homophonically — A could be sent as the word for ant, apple or axe — layered over a codebook of several hundred military terms. It was not "speaking Navajo."
Joe Kieyoomia, a Navajo speaker held as a prisoner of war, was forced to listen to intercepted transmissions. He recognised his own language and could not read a word of it, because he did not have the codebook. He was tortured in the course of it. Popular retellings credit the rarity of the language; the record credits the codebook. That is the switch in the left rail, and it is the one that changes nothing.
By modern standards the scheme is a homophonic substitution over a rare language. Contemporary analyses hold that a team pairing a native speaker with trained cryptanalysts would likely have broken it. The Japanese had the speaker and never assembled the team — an operational failure on their side, not cryptographic strength on the Allied side.