Interactive labs
Eight labs that compute real results
None of these is a canned animation. Each runs an actual model in your browser — deterministic, seeded where it matters, and reproducible. Each opens inside the lesson that teaches the reasoning behind it.
The two experiments worth running first. In the BB84 simulator, set Eve to
photon-number splitting with plain BB84 — the attack succeeds and the error rate stays clean. Then
switch the protocol to decoy-state BB84 and run it again. Do the same with detector blinding and the
optical monitor toggle. Together they demonstrate the single most important operational fact about
QKD security: a clean QBER is not evidence of a secure link.