Zeeka saysA zero-knowledge proof convinces a verifier a statement is true while revealing nothing else — resting on three properties: completeness, soundness, and zero-knowledge.
A zero-knowledge proof lets a prover convince a verifier that some statement is true without revealing why it is true. Three properties define it. Completeness: an honest prover always convinces an honest verifier. Soundness: a cheating prover is caught except with negligible probability. Zero-knowledge: the verifier learns nothing beyond the bare fact that the statement holds.
The classic intuition is a cut-and-choose game: each round, a cheater who lacks the secret can fake an answer only half the time, so after n rounds their odds of getting away with it are 2⁻ⁿ — vanishingly small. Zero-knowledge is captured by the idea that the entire transcript could have been simulated without the secret, so it cannot leak it. The demo runs many rounds to show completeness and soundness numerically.
Power-ups you unlock
Completeness: an honest prover always convinces an honest verifier
Soundness: a cheating prover is caught except with negligible probability
Zero-knowledge: the verifier learns nothing beyond the statement being true
Repeated challenge rounds drive the soundness error toward zero (2⁻ⁿ)
The transcript could have been simulated → it leaks no secret
The Collision attacks — common mistakes
Confusing zero-knowledge (leaks nothing) with mere privacy
Assuming one round is enough — soundness needs many rounds or a big field
Thinking a proof of knowledge and a proof of a fact are identical
Believing ZK hides the statement — only the witness is hidden
Boss battleSimulate many challenge rounds and show an honest prover always passes while a cheater passes only with probability 2⁻ʳᵒᵘⁿᵈˢ.